Skip to main content

Full text of "ACGIH: Industrial Ventilation Manual"

See other formats


CERIMAICATE 


By Authority Of 
THE UNITED STATES OF AMERICA 


Legally Binding Document 


By the Authority Vested By Part 5 of the United States Code § 552(a) and 
Part 1 of the Code of Regulations § 51 the attached document has been duly 
INCORPORATED BY REFERENCE and shall be considered legally 
binding upon all citizens and residents of the United States of America. 
HEED THIS NOTICE: Criminal penalties may apply for noncompliance. 


WW 


Document Name: ACGIH: Industrial Ventilation Manual 


CFR Section(s): 42 CFR 52b.12 


Standards Body: American Conference of Governmental Industrial 


Hygienists 


Official Incorporator: 


THE EXECUTIVE DIRECTOR 
OFFICE OF THE FEDERAL REGISTER 
WASHINGTON, D.C. 


A Manual of Recommended Practice 


23rd Edition 


1998 


American Conference of Governmental Industrial Hygienists 
1330 Kemper Meadow Drive 
Cincinnati, Ohio 45240-1634 


Copyright © 1998 
by 


American Conference of Governmental Industrial Hygienists, Inc. 


Previous Editions 


Copyright © 1951, 1952, 1954, 1956, 1958, 1960, 1962, 1964, 1966, 
1968, 1970, 1972, 1974, 1976, 1978, 1980, 1982, 1984, 1986 


by 


Committee on Industrial Ventilation 
American Conference of Governmental Industrial Hygienists 


Ist Edition — 1951 
2nd Edition — 1952 
3rd Edition — 1954 
4th Edition — 1956 
5th Edition — 1958 
6th Edition — 1960 
7th Edition — 1962 
8th Edition — 1964 
9th Edition — 1966 
10th Edition — 1968 
1 1th Edition — 1970 
12th Edition — 1972 


13th Edition — 1974 
{4th Edition — 1976 
1 Sth Edition — 1978 
16th Edition — 1980 
17th Edition — 1982 
18th Edition — 1984 
19th Edition — 1986 
20th Edition — 1988 
21st Edition — 1992 
22nd Edition — 1995 
23rd Edition — Metric —1998 


Third Printing 


ALL RIGHTS RESERVED. No part of this work covered by the copyright hereon may be reproduced or used in any form or by 
any means — graphic, electronic, or mechanical including photocopying, recording, taping, or information storage and retrieval 
systems — without written permission of the publisher. 


Published in the United States of America by 
American Conference of Governmental Industrial Hygienists, Inc. 
1330 Kemper Meadow Drive 
Cincinnati, Ohio 45240-1634 


ISBN: 1-882417-22-4 
Printed in the United States 


CONTENTS 


DEDICATION) pA RAS ee 2 oh Phe Oo peaane aie tern BO ak Sl a tehe ds abe cet a meee a Le vil 
FOREWORD: *. 0 gaa Soh ghee bad Swe we Meth el ee eA Bla Se Se es ten ix 
ACKNOWLEDGMENTS © 2/6 be4 4-5 d-d0dod Bie Eee NE aes Se i ek SD ee a oo Behe dg xi 
DEFINITIONS” -bS.w Sh. bw dle ee a eae OE Se oe LEG Gee Ea ae bay PLAS xiii 
ABBREVIATIONS’: -Srsuech et c's oo bt pean Bond ole ee eae le weg ie Grn aed Baw ee a Re ee XV 
CHAPTER 1 GENERAL PRINCIPLES OF VENTILATION .......0..0.0202020 0002 eee eee 1-1 
Tel. Antroduction:. 2.29 ek ee Be AL Eo ever doe BO ee Soe Sn ge Ae ch ick & 1-2 

1.2:4 :Supply:Systems~ 9.2 2\6 @ a. a nee ld Oe an Ga Soe be a ee th be got Bak bux. 1-2 

13> SExhaust Systems’ 13 ac2 8 Sata“ te Bn Shwe ofa k Sihurtra te oi ais Pele to toe phe as aoe ans 1-2 

4) “Basic Definitions :2 isa sales ek Gem eee Gop hoe Oe Se See eS Be Pe ee PS 1-3 

15. -Principles‘of-AinFlOwW® ¢ ..pa4 vo ek ee Ok se ep ea ee ot ee tl he 1-4 

1.6 Acceleration of Air and Hood Entry Losses... 2... ee 1-6 

7 Duct-Losseés® 2.4 .650 8 doe ee ee A oe es oh an Ree he Gree ge 8 1-7 

1.8 | Multiple-Hood Exhaust Systems 2... 2... 2 ee 1-9 

1.9 Air Flow Characteristics of Blowing and Exhausting.. 2... 0.0. .2.-000.0.0000004 1-10 

References! 2 ods deh Pe 2 le Me Bet £4 ee eb Se Mk es SEE oe Oe 1-10 

CHAPTER 2 GENERAL INDUSTRIAL VENTILATION. 2.0.0. 2 2-1 
2c TintrOduetion. ia nd ones, aa PR woe arid BOR IE Ses See Nae aS eee 2-2 

2.2 Dilution Ventilation Principles . 2... 2... ee 2-2 

2.3 Dilution Ventilation for Health 2.2.2... 2. 2-2 

2.4 Mixtures—Dilution Ventilation for Health .. 2.0... ee 2-6 

2.5 Dilution Ventilation for Fire and Explosion... 2... 2 2. ee 2-7 

2.6 Fire Dilution Ventilation for Mixtures... 00 2-8 

2.7. ‘Ventilation for Heat Control... 2. 2-8 

2.8. Heat Balance and.Exchanige =: & 222s 2 ¥ ad i dai eek ee bee Yaw weed ae wed 2-8 

2.9 Adaptive Mechanism of the Body .. 2... 0. ee 2-9 

2:10 Acchimatization: »,.. 0-3) 4k Shao ye awn ene! Bae es aad, Peow a a heed tee Gated 2-10 

21k * “Acute-Heat Disorders: uence So Melk neh che etek eG Woe ag. lhe 2 ke de 8 2-10 

2.12 Assessment of Heat Stress and Heat Strain... 2. ee 2-11 

2:13: Worker Protection. 5.4a:.0'6! o.0°S ho Bed ele bra Bara ed oA Mae den end Pe 2-13 

2.44 Ventilation:Control:.” 3.008 3. ese aca wk od eb eat Pea. Me eS A ee, & Ge ae 2-13 

2.13 “Ventilation Systems’ a. .ogo6 gk kb eye de dese Se Gotan la Mh ia Gay abs 2-13 

216° Velocity Cooling. cs end ak eke hl tne Boe bad WON sng ww Pe wo eae 2-15 

2:47" Radiant Heat:Control: 18% Sse Soe BAe ey ye i RA wa ea ee E eRe ee Soc 2-15 

2.18 Protective Suits for Short Exposures 2.2... 0. ee 2-16 

2.19 Respiratory Heat Exchangers .. 2... ee 2-16 

2:20. Refrigerated Suits): j3,.2:30 4 4p oe heed a eee deg OED ok Seen ok 2-16 

ZIV. SENCIOSUTES. 2: a4cis4G: dre Ye Ase hb tee Ale doe a ea a AG eG OR Ea ee te es 2-16 

222 1 ANSUVAION ee 5 ered Sein ees an Bete) ea oe Bee Be Mild Sc Ee ad Rae tele 2-16 

References: ste ara len Boe BS eno Rae he ar ee et ace CR Ua aed 2-17 

CHAPTER 3 LOCAL EXHAUST HOODS.......0...0.0.0.0 00.002. 202 cc ee 3-1 
Sb Antroductiony: 6. ee Gwe eo es a ek Ra en Bote Pee A A ee ae sb kG 3-2 

3.2 Contaminant Characteristics. 2... 2 ee 3-2 

33% “IMOGd Types? "2, ates utr Pe ek te Ee Sree Bale RO Ea ens oa ee A pce ie'ek PAO 3-2 

3.4 Hood Design Factors... 2.2... ee 3-2 

3:5° - Hood. Losses) os. iu bee eee Pee a Pee o4 6 he eee eee oe Fe 3-15 


iii 


iv 


CHAPTER 4 


CHAPTER 5 


CHAPTER 6 


CHAPTER 7 


Industrial Ventilation 


3.6 Minimum Duct Velocity .. 2... .0.00 000200022 ee 3-18 
3.7 Special Hood Requirements .......0...2. 2020-000 0000 eee 3-18 
3.8  Push-Pull Ventilation... .......0020 002000000002 eee ee 3-19 
3:9". THOUPrOCeSSES? 2.43 4 ee eee eee ee aie dy ea Re tia eden 3-21 

RGPEreNCES 54. he thy a eee Ae eotice et We eel hg, ale lena le Oe een hla) a Ye aes 3-23 
AIR: CLEANING: DEVICES 3... 2) s-4.00. ah Goh Sh Sie bal a pewea oa Radek doe WS bors, o 4-1 
AA HIMtOdUCHON 20%. eG eo eA heels he ts Sa ee os ee a has Se eS Be 4-2 
4.2 Selection of Dust Collection Equipment ...........2.2..2.2.. 020220000000. 4-2 
4.3-. \Dust-Gollector Lypes:2-0 3 2s SAS ee EOS EES eae eRe RAN ae Se 4-3 
4.4 Additional Aids in Dust Collector Selection... 2... .0..0.0022200 002.0000 ee eee 4-22 
45 Control of Mist, Gas, and Vapor Contaminants .............0.0..0 2.000000. 4-22 
4.6 Gaseous Contaminant Collectors... 2. 2 2. 4-25 
4.7, “WnitiCollectors ase. go % hk doce eee ew ac beh ba ce Roe Mien ee eS 4-25 
4.8 Dust Collecting Equipment Cost... 0... 0 ee 4-25 
4.9 Selection of Air Filtration Equipment .... 0... .002.0.00200 020200000002. eee 4-28 
4.10 Radioactive and High Toxicity Operations... 2... ee 4-33 
400) Explosion; Venting: 22%. sie Soke 2 het o SLA GP gee en ted eh a Gea Stet a Bw eS 4-33 

References: eck ee Oe See Sy he is & DA eats etl wy Sve eas. 2 4-34 
EXHAUST SYSTEM DESIGN PROCEDURE... 2... .0.0-020000 20 ee 5-] 
Sul. Wntroductonis 4-024 gc & oS eee Bi Pe eG See Pew So 6b Ar ep Pe ce 5-2 
5.2: “Preliminary: Steps. 4 je. doe we poe Sie RE wae ee Bae ee ee ee Fs 5-2 
533°. sDesigni Procedure ‘on-2 5 sets hie gene eh ths abe aS, os ae debe GOS Golan, DE Sew. Pow gt eee ta 5-2 
5:4. Duct-Segment:Calculations!::: ¢ ec..¢-5 8 eee ae a a Re a Ee a ee ee 5-3 
5.5 Distribution of AirFlow .. 2... 0.0. 5-4 
5:6, cAids:-to: Calculation’: oe 33.c dar sab GN eS etew ale he a A ha eee 5-11 
§.7 --Plenum:/Exhaust Systems: ..2.5.9 0205. ue ol A Sch e Bhilai le eae we ee 5-1] 
5:8: - “Fan-Pressure:Caloulations ioc... 320s ep. Hoa ee ded en Ge een, aE a eo 5-11 
5.9 Corrections for Velocity Changes 2.2... ee 5-12 
5/10'.: Sample: System: Design 3:1. et. 0d Agee Pe ee eA A Sk ye Sn eel ys dak 5-13 
5.11 Different Duct Material Friction Losses 2... 2... 0.02 2 ee 5-13 
5.12 Friction Loss for Non-Circular Ducts. 2... ee 5-13 
5.13 Corrections for Non-Standard Density... 0... ee 5-15 
5.14: Air Cleaning Equipment: @.. (2s 3 oat Seoul Sod oe He eldest Ra 5-32 
Slo sEVaSe: Chun etees ters 8 eee he A ls ed tepals eh a od San dl aig eet rete toute Me Sry 5-32 
9:16: “ExhaustS tack Outlets: s, 3. ace te ee a aus, Pca ee ee ee ts et Se a ca, le BEA 5-33 
S17 Air Bleed2lnss.4.¢0 22404404 b24 0 baie d ethane 6628 44 Pa gee ve dees 5-35 
5.18 Optimum Economic Velocity... 2... 02 5-35 
5.19 Construction Guidelines for Local Exhaust Systems... 2... 0.00 2 ee ee $-35 

References enc: fist Bi ae eth Gk Ate, needing Bike dea Yoo hiadie Bae he Bo 5-38 
PAINS} ii 204-2 ict eet dee ee id 1s hts Bec el ae Ale ht od eo ih dtd ie ee ed ae de Se 6-1 
Gil. Introductions (ses see se ceais & Ee ea See en ee ce Be Be eRe oe eA 6-2 
6:2. .Basic'Définitions 05.4% 3 fF oe Sa Sebo wei ble ow hoes Eee oho tes Fads 6-2 
6:3" Fan Selection: sc; 029 ee dye span d Atle tee ebb aos eb Ge aed Cae Wed a eth aS 6-6 
6.4 Fan Installation and Maintenance... 2... 2. 2 ee 6-21 

References. sor cis Be dene ali i had a es eR EO oe Ew Se ee is ole 6-25 
REPLACEMENT AND RECIRCULATED AIR ........0.0...-.......- As Axe an ete 7-1 
7A. -InttoductiOn: 20-2 e.2 ce Pee ee a eo ey eH Mahe deb Re ee eae te de 7-2 
7.2.- ‘Replacement Am”. «2 9 200 et ecis: faede Epa ee A ee aa, aoe a eS 7-2 
7.3. Replacement Air Distribution. ©... 2. ee 7-4 
7:4: Replacement Air Flow Rate-. 3.04.4. 5 2 oe be a eG ee ee ee ed 7-5 


75) Room’ Pressure..3: g.3)t.ce 63) hod) egw eS oh alan fle dS A oD 

7.6 Environmental Control... 2... 0. ee 

7.7 Environmental Control Air Flow Rate... 2... 0. ee 

PBS CAM CHANGES (fe se We So hl ee eh ats eo BA eM oe es 

7.9 Air Supply Temperatures 2... 2... ..202.20.0.002.02 22.2. 20022 0000. 

7.10 Air Supply Vs. Plant Heating Costs... ... le ie 1 es Ee gee ae a Bae 8s 

7.11 Replacement Air Heating Equipment ...............00-..2000. 

7.12 Cost of Heating Replacement Air... 2... 2. 0020.02020.2.2.22.020004 

7.13. ~Air Consérvation=s: 60 1 bd ated oe how hee eae Bea as as 

7.14 Evaluation of Employee Exposure Levels... 2... 0.20.20... 0200-00.. 

References; 23.325 sed ind Sah ohh ik eR Ste de) ee See a Ware eee 

CHAPTER 8 VENTILATION ASPECTS OF INDOOR AIR QUALITY ................ 
S20 Tntroduction. a. a! 3,464 -44¢o05 bk & Ale ee -3 4S oa RG bie ware & 

8.2 Dilution Ventilation for Indoor Air Quality... .......0......-.2.00. 

8.3. HVAC Components and System Types... 2... 2... 00... 20000-02002 

8.4 HVAC Components, Functions and Malfunctions .............--.. 

8.5 HVAC Component Survey Outline... 2... 0200.20.20... 00220008 

References ioe seasoned on tk de ee ON Sr eaten Pe NL et tae 

CHAPTER 9 TESTING OF VENTILATION SYSTEMS ..........-..2..0.02.2 0220004 
9:1 Introdtiction® 36/5, Bas ea ek ee BA eo Ae AAAS bs eae ER 

9.2 Measurements of Volumetric Flow Rate... 2... eee 

9.3 Calibration of Air Measuring Instruments... 2... ....0.0.0-.00004 

9.4 Pressure Measurement... 2.2... 2. ee 

9.5 Pitot Traverse Method .. 2... 2... ee 

9.6 Corrections for Non-Standard Conditions... .........0......00.. 

9.7 Check-out Procedures 2. 22 ee 

References: fas. aoe ate he a ae eal dad haleget.a he Bb ts ee 

CHAPTER 10 SPECIFIC OPERATIONS ......0.0..0200 000000 cee eee 
BIBLIOGRAPHY? © 4a. 824 gc$cdcp tagce: waendth het love eect inal tes Moet, oe SOS Pc Rew 
APPENDICES) ait dipaccis i Gate Boeraty a tik WA Be Math a ot a ges MN ee ad Ae a de gh Bb 


A Threshold Limit Values for Chemical Substances in the Work 
Environment with Intended Changes for 1996-1997 
B Physical Constants/Conversion Factors 


INDEX oA eb ey oe a oe ls Shh Gas tebe pa aad hed aie ity antes 


Contents 


Vv 


This Edition is Dedicated in Memory of 
KNOWLTON J. CAPLAN, PE, CiH, CSP 
June 23, 1919 - April 11, 1997 


Knowlton J. Caplan, while at the Division of Occupational 
Health, Michigan Department of Health, supervised the 
preparation of a field manual on industrial ventilation. That 
manual became the basis of the first edition of /ndustrial 
Ventilation in 1951. For the next forty-six years, the Ventila- 
tion Committee has felt Caplan’s presence as we published 
the “Vent Manual.” This 23" edition is no different. Although 
“Cap” has not been an active member of the Committee for 
the past eleven years, his presence was felt at almost every 
meeting. Frequently we punctuated discussions with a quota- 
tion from Cap or a reference to one of his published works. 
Because of his influence, we proudly dedicate this edition to 
Knowlton J. Caplan. 


During his 50-year career, Cap was a pioneer in the fields 
of industrial hygiene, industrial ventilation, and air pollution 
control. He conducted basic research on cyclone and fabric 
filter dust collectors and holds several patents for these de- 
vices. As an associate professor in the public health depart- 
ment of the University of Minnesota, Cap advised numerous 
Master’s degree students in industrial hygiene, occupational 
health and air pollution control. He has been an instructor at 
the industrial ventilation conferences at Michigan State Uni- 
versity (thirty years) and the University of Washington (ten 
years). As an author of more than 70 technical papers, he was 
a frequent presenter at the American Industrial Hygiene Con- 
ference and the American Society of Heating Refrigeration 
and Air Conditioning Engineers (ASHRAE) meeting. In ad- 
dition he wrote chapters in Air Pollution by Stern, Industrial 
Hygiene and Toxicology by Patty, Uranium Production Tech- 
nology by Harrington and Rueble, and was the Associate 
Editor of Industrial Hygiene Aspects of Plant Operations: 
Volume 3 - Engineering Considerations in Equipment Selec- 


Vii 


tion, Layout and Building Design by Crawley and Crawley. 


Besides his innovative ventilation design, Cap developed a 
method for testing laboratory fume hoods which won the Best 
Paper of the Year award of the Michigan Industrial Hygiene 
Society in 1982, which later became the basis for the 
ASHRAE Standard 110-1995, “Method of Testing Labora- 
tory Fume Hood Performance.” Cap was a significant partici- 
pant in the development of the ANSI Standard Z9.5-1992, 
“American National Standard for Laboratory Ventilation.” He 
was the first to employ “clean air islands” to supplement local 
exhaust ventilation where necessary. 


Cap has been active in several societies: ACGIH (Commit- 
tee on Industrial Ventilation), Air Pollution Control Associa- 
tion (Committee on Dust, Fume, and Mist Control), American 
Industrial Hygiene Association (Board of Directors, Air Pol- 
lution Control Committee), ASHRAE ( Industrial Ventila- 
tion, Industrial Process Air Cleaning), American National 
Standards Committee (Air Pollution Committee, Health and 
Safety Committee), American Board of Industrial Hygiene. 


Cap was born in St. Louis, Missouri. He earned his bache- 
lor’s and master’s degree in chemical engineering from Wash- 
ington University in the 1940s. He served in the 
Commissioned Corps of the U.S. Public Health Services. He 
worked as a chemical engineer and ventilation engineer at 
Ralston Purina Company and Mallinckrodt Chemical, Ura- 
nium Division. He also worked for the St. Louis County 
Health Department and the Michigan Department of Health 
as an industrial hygienist. In addition, Cap did consulting 
work, primarily as a ventilation engineer for Industrial Health 
Engineering Associates (co-founder), Pace Incorporated, and 
Rust Environment and Infrastructure. 


FOREWORD 


Industrial Ventilation: A Manual of Recommended Practice 
is the outgrowth of years of experience by members of the 
ACGIH Industrial Ventilation Committee members and a com- 
pilation of research data and information on design, mainte- 
nance, and evaluation of industrial exhaust ventilation systems. 
The Manual attempts to present a logical method of designing 
and testing these systems. It has found wide acceptance as a guide 
for official agencies, as a standard for industrial ventilation 
designers, and as a textbook for industrial hygiene courses. 


The Manual is not intended to be used as law, but rather as a 
guide. Because of new information on industrial ventilation 
becoming available through research projects, reports from en- 
gineers, and articles in various periodicals and journals, review 
and revision of each section of the Manual is an ongoing Com- 
mittee project. The Manual is available as a hardbound book and 
on CD-ROM. In a constant effort to present the latest techniques 
and data, the Committee desires, welcomes, and actively seeks 
comments and suggestions on the accuracy and adequacy of the 
information presented herein. 


In this 23rd edition, the Committee has made a number of 
minor revisions. Chapter 5 includes updated duct calculation 
sheets designed to aid in calculations. The “3 eye” duct friction 
charts have been replaced with tables to permit easier determi- 
nation of the duct friction factor. The metric supplement has been 


deleted and the Committee has developed a separate metric 
manual. 


This publication is designed to present accurate and 
authoritative information with regard to the subject matter 
covered. It is distributed with the understanding that nei- 
ther the Committee nor its members collectively or indi- 
vidually assume any responsibility for any inadvertent 
misinformation, omissions, or for the results in the use of 
this publication. 


COMMITTEE ON INDUSTRIAL VENTILATION 


R.T. Hughes, NIOSH, Ohio, Chair 

A.G. Apol, FEOH, Washington 

W.M. Cleary, Retired, Michigan 

M.T. Davidson, The New York Blower Co., Indiana 
T.N. Do, NFESC, California 

Mrs. Norma Donovan, Editorial Consultant 

S.E. Guffey, U. of Washington, Washington 

G.S. Knutson, Knutson Ventilation Consultants, Minnesota 
G. Lanham, KBD/Technic, Ohio 

kK. Mead, NIOSH, Ohio 

K.M. Paulson, NFESC, California 

O.P. Petrey, Phoenix Process Equipment Co., Kentucky 
A.L. Twombly, Pfeiffer Engineering Co. Inc., Kentucky 


ACKNOWLEDGMENTS 


Industrial Ventilation is a true Committee effort. It brings 
into focus in one source useful, practical ventilation data from 
all parts of the country. The Committee membership of indus- 
trial ventilation and industrial hygiene engineers represents a 
diversity of experience and interest that ensures a well- 
rounded, cooperative effort. 


From the First Edition in 1951, this effort has been success- 
ful as witnessed by the acceptance of the "Ventilation Man- 
ual" throughout industry, by governmental agencies, and as a 
worldwide reference and text. 


The present Committee is grateful for the faith and firm 
foundation provided by past Committees and members listed 
below. Special acknowledgment is made to the Division of 


Occupational Health, Michigan Department of Health, for 
contributing their original field manual which was the basis 
of the First Edition, and to Mr. Knowlton J. Caplan who 
supervised the preparation of that manual. 


The Committee is grateful also to those consultants who 
have contributed so greatly to the preparation of this and 
previous editions of /ndustrial Ventilation and to Mrs. Norma 
Donovan, Secretary to the Committee, for her untiring zeal in 
our efforts. 


To many other individuals and agencies who have made 
specific contributions and have provided support, sugges- 
tions, and constructive criticism, our special thanks. 


COMMITTEE ON INDUSTRIAL VENTILATION 


Previous Members 


A.G. Apol, 1984—present 

H. Ayer, 1962-1966 

R.E. Bales, 1954-1960 

J. Baliff, 1950-1956; Chair, 1954-1956 

J.T. Barnhart, Consultant, 1986-1990 

J.C. Barrett, 1956-1976; Chair, 1960-1968 

J.L. Beltran, 1964-1966 

D. Bonn, Consultant, 1958-1968 

D.J. Burton, 1988-1970 

K.J. Caplan, 1974-1978; Consultant, 1980-1986 
W.M. Cleary. 1978-1993; Consultant, 1993-present; Chair, 
1978-1984 

L. Dickie, 1984-1994; Consultant 1968-1984 
B. Feiner, 1956-1968 

M. Franklin, 1991-1994 

S.E. Guffey, 1984-present 

G.M. Hama, 1950-1984; Chair, 1956-1960 

R.P, Hibbard, 1968-1994 

R.T. Hughes, 1976-present; Chair, 1989-present 
H.S. Jordan, 1960-1962 

J. Kane, Consultant, 1950-1952 


XI 


J. Kayse, Consultant, 1956-1958 

J.F. Keppler, 1950-1954, 1958-1960 

G.W. Knutson, Consultant, 1986-present 
J.J. Loeffler, 1980-1995; Chair, 1984-1989 
J. Lumsden, 1962-1968 

J.R. Lynch, 1966-1976 

G. Michaelson, 1958-1960 

K.M. Morse, 1950-1951; Chair, 1950-1951 
R.T. Page, 1954-1956 

K.M. Paulson, 1991-present 

O.P. Petrey, Consultant, 1978-present 

G.S. Rajhans, 1978-1995 

K.E. Robinson, 1950-1954; Chair, 1952-1954 
A. Salazar, 1952-1954 

E.L. Schall, 1956-1958 

M.M. Schuman, 1962-1994; Chair, 1968-1978 
J.C. Soet, 1950-1960 

A.L. Twombly, Consultant, 1986-present 
J. Willis, Consultant, 1952-1956 

R. Wolle, 1966-1974 

J.A. Wunderle, 1960-1964 


DEFINITIONS 


Aerosol: An assemblage of small particles, solid or liquid, 
suspended in air. The diameter of the particles may vary 
from 100 microns down to 0.01 micron or less, e.g., 
dust, fog, smoke. 


Air Cleaner: A device designed for the purpose of remov- 
ing atmospheric airborne impurities such as dusts, 
gases, vapors, fumes, and smoke. (Air cleaners include 
air washers, air filters, eletrostatic precipitators, and 
charcoal filters.) 


Air Filter: An air cleaning device to remove light particu- 
late loadings from normal atmospheric air before intro- 
duction into the building. Usual range: loadings up to 
3 grains per thousand cubic feet (0.003 grains per cubic 
foot). Note: Atmospheric air in heavy industrial areas 
and in-plant air in many industries have higher loadings 
than this, and dust collectors are then indicated for 
proper air cleaning. 


Air Horsepower: The theoretical horsepower required to 
drive a fan if there were no loses in the fan, that is, if its 
efficiency were 100 percent. 


Air, Standard: Dry air at 70 F and 29.92 in (Hg) barometer. 
This is substantially equivalent to 0.075 Ib/ft3. Specific 
heat of dry air = 0.24 btu/Ib/F. 


Aspect Ratio: The ratio of the width to the length; AR = 
WIL. 


Aspect Ratio of an Elbow: The width (W) along the axis 
of the bend divided by depth (D) in plane of bend; AR 
=W/D. 


Blast Gate: Sliding damper. 


Blow (throw): In air distribution, the distance an air stream 
travels from an outlet to a position at which air motion 
along the axis reduces to a velocity of 50 fpm. For unit 
heaters, the distance an air stream travels from a heater 
without a perceptible rise due to temperature difference 
and loss of velocity. 


Brake Horsepower. The horsepower actually required to 
drive a fan. This includes the energy losses in the fan 
and can be determined only by actual test of the fan. 
(This does not include the drive losses between motor 
and fan.) 


Capture Velocity: The air velocity at any point in front of 
the hood or at the hood opening necessary to overcome 
opposing air currents and to capture the contaminated 
air at that point by causing it to flow into the hood. 


xiii 


Coefficient of Entry: The actual rate of flow caused by a 
given hood static pressure compared to the theoretical 
flow which would result if the static pressure could be 
converted to velocity pressure with 100 percent effi- 
ciency. It is the ratio of actual to theoretical flow. 


Comfort Zone (Average): The range of effective tempera- 
tures over which the majority (50% or more) of adults 
feel comfortable. 


Convection: The motion resulting in a fluid from the 
differences in density and the action of gravity. In heat 
transmission, this meaning has been extended to in- 
clude both forced and natural motion or circulation. 


Density: The ratio of the mass of a specimen of a substance 
to the volume of the specimen. The mass of a unit 
volume of a substance. When weight can be used with- 
out confusion, as synonymous with mass, density is the 
weight of a unit volume of a substance. 


Density Factor: The ratio of actual air density to density 
of standard air. The product of the density factor and 
the density of standard air (0.075 lb/ft3) will give the 
actual air density in pounds per cubic foot; d x 0.075 = 
actual density of air, Ibs/ft3. 


Dust: Small solid particles created by the breaking up of 
larger particles by processes crushing, grinding, drill- 
ing, explosions, etc. Dust particles already in existence 
in amixture of materials may escape into the air through 
such operations as shoveling, conveying, screening, 
sweeping, etc. 


Dust Collector: An air cleaning device to remove heavy 
particulate loadings from exhaust systems before dis- 
charge to outdoors. Usual range: loadings 0.003 grains 
per cubic foot and higher. 


Entry Loss: Loss in pressure caused by air flowing into a 
duct or hood (inches H,O). 


Fumes: Small, solid particles formed by the condensation 
of vapors of solid materials. 


Gases: Formless fluids which tend to occupy an entire 
space uniformly at ordinary temperatures and pres- 
sures. 


Gravity, Specific: The ratio of the mass of a unit volume 
of a substance to the mass of the same volume of a 
standard substance at a standard temperature. Water at 
39.2 F is the standard substance usually referred to. For 
gases, dry air, at the same temperature and pressure as 
the gas, is often taken as the standard substance. 


xiv Industrial Ventilation 


Hood: A shaped inlet designed to capture contaminated air 
and conduct it into the exhaust duct system. 


Humidity, Absolute. The weight of water vapor per unit 
volume, pounds per cubic foot or grams per cubic 
centimeter. 


Humidity, Relative: The ratio of the actual partial pressure 
of the water vapor in a space to the saturation pressure 
of pure water at the same temperature. 


Inch of Water: A unit of pressure equal to the pressure 
exerted by a column of liquid water one inch high at a 
standard temperature. 


Lower Explosive Limit: The lower limit of flammability or 
explosibility of a gas or vapor at ordinary ambient 
temperatures expressed in percent of the gas or vapor in 
air by volume. This limit is assumed constant for tem- 
peratures up to 250 F. Above these temperatures, it 
should be decreased by a factor of 0.7 since explosibility 
increases with higher temperatures. 


Manometer: An instrument for measuring pressure; essen- 
tially a U-tube partially filled with a liquid, usually 
water, mercury or a light oil, so constructed that the 
amount of displacement of the liquid indicates the pres- 
sure being exerted on the instrument. 


Micron: A unit of length, the thousandth part of 1 mm or 
the millionth of a meter (approximately 1/25,000 of an 
inch). 

Minimum Design Duct Velocity: Minimum air velocity 
required to move the particulates in the air stream, fpm. 


Mists: Small droplets of materials that are ordinarily liquid 
at normal temperature and pressure. 


Plenum: Pressure equalizing chamber. 


Pressure, Static: The potential pressure exerted in all di- 
rections by a fluid at rest. For a fluid in motion, it is 
measured in a direction normal to the direction of flow. 
Usually expressed in inches water gauge when dealing 
with air. (The tendency to either burst or collapse the 


pipe.) 
Pressure, Total: The algebraic sum of the velocity pressure 
and the static pressure (with due regard to sign). 


Pressure, Vapor: The pressure exerted by a vapor. If a 
vapor is kept in confinement over its liquid so that the 
vapor can accumulate above the liquid, the temperature 
being held constant, the vapor pressure approaches a 
fixed limit called the maximum or saturated vapor pres- 


sure, dependent only on the temperature and the liquid. 
The term vapor pressure is sometimes used as synony- 
mous with saturated vapor pressure. 


Pressure, Velocity: The kinetic pressure in the direction of 
flow necessary to cause a fluid at rest to flow at a given 
velocity. Usually expressed in inches water gauge. 


Radiation, Thermal (Heat) Radiation: The transmission of 
energy by means of electromagnetic waves of very long 
wave length. Radiant energy of any wave length may, 
when absorbed, become thermal energy and result in an 
increase in the temperature of the absorbing body. 


Replacement Air: A ventilation term used to indicate the 
volume of controlled outdoor air supplied to a building 
to replace air being exhausted. 


Slot Velocity: Linear flow rate of contaminated air through 
slot, fpm. 


Smoke: Anair suspension (aerosol) of particles, usually but 
not necessarily solid, often originating in a solid nu- 
cleus, formed from combustion or sublimation. 


Temperature, Effective: An arbitrary index which com- 
bines into a single value the effect of temperature, 
humidity, and air movement on the sensation of warmth 
or cold felt by the human body. The numerical value is 
that of the temperature of still, saturated air which 
would induce an identical sensation. 


Temperature, Wet-Bulb: Thermodynamic wet-bulb tem- 
perature is the temperature at which liquid or solid 
water, by evaporating into air, can bring the air to 
saturation adiabatically at the same temperature. Wet- 
bulb temperature (without qualification) is the tempera- 
ture indicated by a wet-bulb psychrometer constructed 
and used according to specifications. 


Threshold Limit Values (TLVs): The values for airborne 
toxic materials which are to be used as guides in the 
control of health hazards and represent time-weighted 
concentrations to which nearly all workers may be 
exposed 8 hours per day over extended periods of time 
without adverse effects (see Appendix). 


Transport (Conveying) Velocity: See Minimum Design 
Duct Velocity. 


Vapor: The gaseous form of substances which are nor- 
mally in the solid or liquid state and which can be 
changed to these states either by increasing the pressure 
or decreasing the temperature. 


ABBREVIATIONS 


Ef 2 A dui ABE Au oe inh fee tee Pe OO Ret de Se SS area 
BCH seed eee ee oe flow rate at actual condition 
APS * Sig eed a3 le adhere 8 ae os air horsepower 
ARS ct be ss Beanies Bk aspect ratio 
Pooky gee inane! bh ak ae eg ee oas A phe ead Slot area 
Bick asa PS BS oo ook ee oe barometric pressure 
GHP: Mecoree Beta hoe: ee ASS brake horsepower 
DADs sede 8 Seah hae brake horsepower, actual 
bhpe. ve ee WOE A brake horsepower, standard air 
Dt.” te te degregeenr delay na hree ie Gree British thermal unit 
btthcc sc. eo eee OS SE ees btu/hr 
Cer ok een Dd i Soba Bae Gos coefficient of entry 
CIM. -jatechs Wis a eee o cubic feet per minute 
CLR eo Sand LBA hey centerline radius 
Doe id Se Reet Apa th A ae aes diameter 
is Sh Ahk Ge he ees oe ade ae get density factor 
EET. sceesin taba oy ong Amar as aes es effective temperature 
PAs seh talk RA eas degree, Fahrenheit 
Bijele foe eae eS duct entry loss coefficient 
Pajcaperdteves d.6: Mt apie ee eta te elbow loss coefficient 
Fa eect bo ate te ke Hewes Go dah entry loss coefficient 
PDI <b 5 RE tee hs ee Pea feos Ge 2 feet per minute 
EPSPs gee hin @ te Ae EAS Ee ars feet per second 
Fey tahoe he Ee es slot loss coefficient 
TET partectuearbete Miwa tes seh ath soc hee choad square foot 
fk. ea AS EE SS laa e eh ws cubic foot 
Be yh cae Bah ou & hie Re gravitational force, ft/sec/sec 
SPM G dol any Pei at eh ee gallons per minute 
OT 5 dense Se! of Se ON Toes Wale, een hace Sy ess ele oe grains 
lige Bie! eb R LES RASS ee duct entry loss 
Wiis ema cote Le elt cao Ml ei overall hood entry loss 
Wad tise ee SAG BAe ate eae ES elbow loss 
Pigg ca eh ral og ie Bee Sey A A oh pe a ie a entry loss 
HEPA ....... high-efficiency particulate air filters 
Apidae hate tl Be esta BS loss in straight duct run 
He eee ty 2S Bk ee duct loss coefficient 
Ap? sScee yas Bk eet ou ee elk Nb A horsepower 
HE ee ok eS ei aie Be She io auieal @ oa hour 
Ne ee etd elt Ge aha td slot or opening entry loss 
VAS Aegis 2. coed te Wine Ad RE a Se at AB A ES eee inch 


XV 


1 8.8, Ske Br oe od ADS ages om square inch 
TWEE rs che opted nes i. A me es ae Beats inches water gauge 
IBes 9 glare due debe meted Oh Genes & pound 
WOT 2 o8h ge. cach ech a dak eee ee AA SE ah ag! pound mass 
LEL* s&s afeu 8s piece hele se lower explosive limit 
MEF $ fog aren ek Ge BARRY mechanical efficiency 
MG as Fed a tubal Bada s et eee anges milligram 
MING Agsioe tee ie ate, Aang Bl ses minute 
TANS 0%, 4 4 atate a Sia iP Sa, poh es millimeter 
MRT a can Rod ee ee mean radiant temperature 
MW. org: ds Ae a ae ahs ok molecular weight 
Dh ale Saeed coi ae Ae 42 4 density of air in Ib/ft? 
DPM) 3.6) win Poh SAE SB hae parts per million 
PSE: 205.4 a aco dye lo BL ae WAP ee pounds per square inch 
PWReiid te Go Se ee Ae ead le id power 
ite FecaQreck be Sms Sic ds OE Be aoe flow rate in cfm 
Onn fog ata e ey corrected flow rate at a junction 
Re 2 ys Sera toe atte te A ¥ pie S Sod degree, Rankin 
RES 2 5 eek at ek ale fate oe relative humidity 
PPM 2. 5h Rak ds ok Ee revolutions per minute 
SCHM: sf ewido a 0k flow rate at standard condition 
SEPM ace setae hee a surface feet per minute 
SP Blt ioe any he tas don pare co specific gravity 
SP ses! ee sertene Gabe ea es & Searcy static pressure 
OP pie Sok higher static pressure at junction of 2 ducts 
SPke: tite ee al Gl ee GO ew BA hood static pressure 
Det wate le Se ROA Sp, system handling standard air 
SDP 8.5 teeta standard temperature and pressure 
GN? o> ite dei k ee ar etn the Threshold Limit Value 
TP aires ict A a Be oe BRR Gh on A total pressure 
Mi 2 Sidh Tints A ee ne Ak th Wate, Boag velocity, fpm 
Vay Bead i tech ey oe Se Spee, dikateoe De duct velocity 
VPS 2 Soke dc deen Byte aye Selec eed velocity pressure 
WP av eriraye dk fd) ee ate duct velocity pressure 
MP oe sg Reels See ee ee resultant velocity power 
MiP Bch ts Fas ti bya Ble aly Wit slot velocity pressure 
Vid daskvg pik chet, de dod oe Re es Slot velocity 
Vi as Ue ATAU wg Ae Wisede eS duct transport velocity 
We so ska feos Bie Gap TE ay SP ead Gi cin de eae ce aha watt 


Chapter 1 
GENERAL PRINCIPLES OF VENTILATION 


12) SUPPLY SYSTEMS*..2 i ohare eee ee eat 4 1-2 1.7.1 FrictionLosses......0.-2....-20-. 1-7 
1.3 EXHAUSTSYSTEMS ................. 1-2 1.7.2 FittingLosses ................. 1-9 
14 BASIC DEFINITIONS ................. 1-3 1.8 MULTIPLE-HOOD EXHAUST SYSTEMS ..... 1-9 
1.5 PRINCIPLES OF AIRFLOW. ............. 1-4 19 AIRFLOW CHARACTERISTICS OF BLOWING 
1.6 ACCELERATION OF AIR AND HOOD ENTRY AND EXHAUSTING ... 0.2... 06500006. 1-10 
LOSSES 6. 6s ee ie bee Sse ae ade oe 8 1-6 REFERENCES icine se ede dee at gage @ aoe ak Ac alt Bs 1-10 
Figure 1-1 SP, VP, and TPataPoint............. 1-4 Figure 1-5 Variation of SP, VP, and TP Through a Ventilation 
Figure 1-2 Measurement of SP, VP, and TP in a Pressurized SYStEM & 6-8 oN Aldi ea Ge ee aS Se ER Tan Ge 1-6 
DUG Ey 8 5098 5s oy toe, eck so, ee agit 1-4 Figure 1-6 Moody Diagram... 2... ..0.-.0.0004. 1-8 
Figure 1-3 SP, VP, and TP at Points in a Ventilation System 1-5 Figure 1-7 Blowing Vs. Exhausting ............ 1-10 


Figure 1-4 


Volumetric Flow Rates in Various Situations .. 1-5 


1-2 Industrial Ventilation 


1.1 INTRODUCTION 


The importance of clean uncontaminated air in the indus- 
trial work environment is well known. Modern industry with 
its complexity of operations and processes uses an increasing 
number of chemical compounds and substances, many of 
which are highly toxic. The use of such materials may result 
in particulates, gases, vapors, and/or mists in the workroom 
air in concentrations that exceed safe levels. Heat stress can 
also result in unsafe or uncomfortable work environments. 
Effective, well-designed ventilation offers a solution to these 
problems where worker protection is needed. Ventilation can 
also serve to control odor, moisture, and other undesirable 
environmental conditions. 


The health hazard potential of an airborne substance is 
characterized by the Threshold Limit Value (TLV®). The TLV 
refers to the airborne concentration of a substance and repre- 
sents conditions under which it is believed that nearly all 
workers may be exposed day after day without adverse health 
effects. The time-weighted average (TWA) is defined as the 
time-weighted average concentration for a conventional 8- 
hour workday and a 40-hour workweek which will produce 
no adverse health effects for nearly all workers. The 
TLV-—TWA is usually used to determine a safe exposure level. 
TLVs are published annually by the American Conference of 
Governmental Industrial Hygienists (ACGIH); revisions and 
additions are made regularly as information becomes avail- 
able. Appendix A of this Manual provides the current TLV 
list for chemical substances as of the date of publication. 


Ventilation systems used in industrial plants are of two 
generic types. The SUPPLY system is used to supply air, 
usually tempered, to a work space. The EXHAUST system is 
used to remove the contaminants generated by an operation 
in order to maintain a healthful work environment. 


A complete ventilation program must consider both the 
supply and the exhaust systems. If the overall quantity of air 
exhausted from a work space is greater than the quantity of 
outdoor air supplied to the space, the plant interior will 
experience a lower pressure than the local atmospheric pres- 
sure. This may be desirable when using a dilution ventilation 
system to control or isolate contaminants in a specific area of 
the overall plant. Often, this condition occurs simply because 
local exhaust systems are installed and consideration is not 
given to the corresponding replacement air systems. Air will 
then enter the plant in an uncontrolled manner through cracks, 
walls, windows, and doorways. This typically results in: 1) 
employee discomfort in winter months for those working near 
the plant perimeter, 2) exhaust system performance degrada- 
tion, possibly leading to loss of contaminant control and a 
potential health hazard, and 3) higher heating and cooling costs. 
Chapter 7 of this Manual discusses these points in more detail. 


1.2 SUPPLY SYSTEMS 


Supply systems are used for two purposes: 1) to create a 


comfortable environment in the plant (the HVAC system); 
and 2) to replace air exhausted from the plant (the REPLACE- 
MENT system). Many times, supply and exhaust systems are 
coupled, as in dilution control systems (see Section 1.3 and 
Chapter 2.) 


A well-designed supply system will consist of an air inlet 
section, filters, heating and/or cooling equipment, a fan, ducts, 
and register/grilles for distributing the air within the work 
space. The filters, heating and/or cooling equipment and fan 
are often combined into a complete unit called an airhouse or 
air supply unit. If part of the air supplied by a system is 
recirculated, a RETURN system is used to bring the air back 
to the airhouse. 


1.3 EXHAUST SYSTEMS 


Exhaust ventilation systems are classified in two generic 
groups: 1) the GENERAL exhaust system and 2) the LOCAL 
exhaust system. 


The general exhaust system can be used for heat control 
and/or removal of contaminants generated in a space by 
flushing out a given space with large quantities of air. When 
used for heat control, the air may be tempered and recycled. 
When used for contaminant control (the dilution system), 
enough outdoor air must be mixed with the contaminant so 
that the average concentration is reduced to a safe level. The 
contaminated air is then typically discharged to the atmos- 
phere. A supply system is usually used in conjunction with a 
general exhaust system to replace the air exhausted. 


Dilution ventilation systems are normally used for con- 
taminant control only when local exhaust is impractical, as 
the large quantities of tempered replacement air required to 
offset the air exhausted can lead to high operating costs. 
Chapter 2 describes the basic features of general ventilation 
systems and their application to contaminant and fire hazard 
control. 


Local exhaust ventilation systems operate on the principle 
of capturing a contaminant at or near its source. It is the 
preferred method of control because it is more effective and 
the smaller exhaust flow rate results in lower heating costs 
compared to high flow rate general exhaust requirements. The 
present emphasis on air pollution control stresses the need for 
efficient air cleaning devices on industrial ventilation sys- 
tems, and the smaller flow rates of the local exhaust system 
result in lower costs for air cleaning devices. 


Local exhaust systems are comprised of up to four basic 
elements: the hood(s), the duct system (including the exhaust 
stack and/or recirculation duct), the air cleaning device, and 
the fan. The purpose of the hood is to collect the contaminant 
generated in an air stream directed toward the hood. A duct 
system must then transport the contaminated air to the air 
cleaning device, if present, or to the fan. In the air cleaner, the 
contaminant is removed from the air stream. The fan must 
overcome all the losses due to friction, hood entry, and fittings 


in the system while producing the intended flow rate. The duct 
on the fan outlet usually discharges the air to the atmosphere 
in such a way that it will not be re-entrained by the replace- 
ment and/or HVAC systems. In some situations, the cleaned 
air is returned to the plant. Chapter 7 discusses whether this 
is possible and how it may be accomplished. 


This Manual deals with the design aspects of exhaust 
ventilation systems, but the principles described also apply to 
supply systems. 


1.4 BASIC DEFINITIONS 


The following basic definitions are used to describe air flow 
and will be used extensively in the remainder of the Manual. 


The density (p) of the air is defined as its mass per unit 
volume and is normally expressed in pounds mass per cubic 
foot (bm/ft*). At standard atmospheric pressure (14.7 psia), 
room temperature (70 F) and zero water content, its value is 
normally taken to be 0.075 Ibm/ft?, as calculated from the 
perfect gas equation of state relating pressure, density, and 
temperature: 


p=pRT [1.1] 


where: 
p= the absolute pressure in pounds per square foot 
absolute (psfa) 
p = the density, 1bm/ft® 
= the gas constant for air and equals 53.35 ft- 
lb/Ilbm-degrees Rankine 
T = the absolute temperature of the air in degrees 
Rankine 
Note that degrees Rankine = degrees Fahrenheit + 459.7. 
From the above equation, density varies inversely with 
temperature when pressure is held constant. Therefore, for 
any dry air situation (see Chapter 5 for moist air calculations), 


pT = (pT) stp 
or 
Tstp 530 
= = 0.075 -—— 
P= Pstp T T [1.2] 


For example, the density of dry air at 250 F would be 


p = 0.075 930 ___ 9.056 Ibm/ft? 


460 + 250 


The volumetric flow rate, many times referred to as "vol- 
umes," is defined as the volume or quantity of air that passes 
a given location per unit of time. It is related to the average 
velocity and the flow cross-sectional area by the equation 


Q=VA [1.3] 


where: 


General Principles of Ventilation 1-3 


Q= volumetric flow rate, cfm 
V= average velocity, fom 
A= cross-sectional area, ft? 


Given any two of these three quantities, the third can readily 
be determined. 


Air or any other fluid will always flow from a region of 
higher total pressure to a region of lower total pressure in the 
absence of work addition (a fan). There are three different but 
mathematically related pressures associated with a moving air 
stream. 


Static pressure (SP) is defined as the pressure in the duct 
that tends to burst or collapse the duct and is expressed in 
inches of water gage ("wg). It is usually measured with a water 
manometer, hence the units. SP can be positive or negative 
with respect to the local atmospheric pressure but must be 
measured perpendicular to the air flow. The holes in the side 
of a Pitot tube (see Figure 9-9) or a small hole carefully drilled 
to avoid internal burrs that disturb the air flow (never 
punched) into the side of a duct will yield SP. 


Velocity pressure (VP) is defined as that pressure required 
to accelerate air from zero velocity to some velocity (V) and 
is proportional to the kinetic energy of the air stream. The 
relationship between V and VP is given by 


v= 1096 |YP. 
p 


or 


2 
Vv 
v=o zoos) (4 


where: 
V= velocity, fpm 
VP = velocity pressure, "wg 


If standard air is assumed to exist in the duct with a density 
of 0.075 1bm/ft’, this equation reduces to 


Vv = 4005/VP 
or 
2 
Vv 
ue (zoos) [1.5] 


VP will only be exerted in the direction of air flow and is 
always positive. Figure I-1 shows graphically the difference 
between SP and VP. 


Total pressure (TP) is defined as the algebraic sum of the 
static and velocity pressures or 


TP =SP+VP [1.6] 


Total pressure can be positive or negative with respect to 


1-4 Industrial Ventilation 


FIGURE 1-1. SP, VP, and TP at a point 


atmospheric pressure and is a measure of the energy content 
of the air stream, always dropping as the flow proceeds 
downstream through a duct. The only place it will rise is 
across the fan. 


Total pressure can be measured with an impact tube point- 
ing directly upstream and connected to a manometer. It will 
vary across a duct due to the change of velocity across a duct 
and therefore single readings of TP will not be representative 
of the energy content. Chapter 9 illustrates procedures for 
measurement of all pressures in a duct system. 


The significance of these pressures can be illustrated as 
follows. Assume a duct segment with both ends sealed was 
pressurized to a static pressure of 0.1 psi above the atmos- 
pheric pressure as shown in Figure 1-2. If a small hole 
(typically 1/16" to 3/32") were drilled into the duct wall and 
connected to one side of a U-tube manometer, the reading 
would be approximately 2.77 "wg. Note the way the left-hand 
manometer is deflected. If the water in the side of the ma- 
nometer exposed to the atmosphere is higher than the water 
level in the side connected to the duct, then the pressure read 
by the gauge is positive (greater than atmospheric). Because 
there is no velocity, the velocity pressure is 0 and SP= TP. A 
probe which faces the flow is called an impact tube and will 


measure TP. In this example, a manometer connected to an 
impact tube (the one on the right) will also read 2.77 "wg. 
Finally, if one side of a manometer were connected to the 
impact tube and the other side were connected to the static 
pressure opening (the center one), the manometer would read 
the difference between the two pressures. As VP = TP — SP, 
a manometer so connected would read VP directly. In this 
example, there is no flow and hence VP = 0 as indicated by 
the lack of manometer deflection. 


If the duct ends were removed and a fan placed midway in 
the duct, the situation might change to the one shown on 
Figure |-3. Upstream of the fan, SP and TP are negative (less 
than atmospheric). This is called the suction side. Down- 
stream of the fan, both SP and TP are positive. This is called 
the pressure side. Regardless of which side of the fan is 
considered, VP is always positive. Note that the direction in 
which the manometers are deflected shows whether SP and 
TP are positive or negative with respect to the local atmos- 
pheric pressure. 


1.5 PRINCIPLES OF AIR FLOW 


Two basic principles of fluid mechanics govern the flow of 
air in industrial ventilation systems: conservation of mass and 
conservation of energy. These are essentially bookkeeping 
laws which state that all mass and all energy must be com- 
pletely accounted for. A coverage of fluid mechanics is not in 
the purview of this manual; reference to any standard fluid 
mechanics textbook will show the derivation of these princi- 
ples. However, it is important to know what simplifying 
assumptions are included in the principles discussed below. 
They include: 


1. Heat transfer effects are neglected. If the temperature 
inside the duct is significantly different than the air 
temperature surrounding the duct, heat transfer will 
occur. This will lead to changes in the duct air tem- 
perature and hence in the volumetric flow rate. 


2. Compressibility effects are neglected. If the overall 
pressure drop from the start of the system to the fan is 
greater than about 20 "wg, then the density will change 


FIGURE 1-2, Measurement of SP, VP, and TP in a pressurized duct 


| 
| 
| 
| 
i 


General Principles of Ventilation 1-5 


PRESSURE SIDE 


3000 form 


uearae —~ 


5000 form 


oP + VP = TP 

-~1.1 + 0.56 = -0.54 
PRESSURES BELOW 

ATMOSPHERIC 


FIGURE 1-3. SP, VP, and TP at points in a ventilation system 


change (see Chapter 5). 


3. The air is assumed to be dry. Water vapor in the air 
stream will lower the air density, and correction for 
this effect, if present, should be made. Chapter 5 
describes the necessary psychrometric analysis. 


4. The weight and volume of the contaminant in the air 
stream is ignored. This is permissible for the contami- 
nant concentrations in typical exhaust ventilation sys- 
tems. For high concentrations of solids or significant 
amounts of gases other than air, corrections for this 
effect should be included. 


Conservation of mass requires that the net change of mass 
flow rate must be zero. If the effects discussed above are 
negligible, then the density will be constant and the net change 
of volumetric flow rate (Q) must be zero. Therefore, the flow 
rate that enters a hood must be the same as the flow rate that 
passes through the duct leading from the hood. At a branch 
entry (converging wye) fitting, the sum of the two flow rates 
that enter the fitting must leave it. At a diverging wye, the 
flow rate entering the wye must equal the sum of the flow 
rates that leave it. Figure 1-4 illustrates these concepts. 


Conservation of energy means that all energy changes must 
be accounted for as air flows from one point to another. In 
terms of the pressures previously defined, this principle can 
be expressed as: 


TP, = TP, +h, 


SP + VP = TP 
0.20 + 0.56 = 0.76 


PRESSURES ABOVE 
ATMOSPHERIC 


/ 5000 fom 


GOs -Q2 


b. Q1+Q2=Q3 


FIGURE 1-4. Volumetric flow rates in various situations. a. Flow through a 
hood; b. Flow through a branch entry 


1-6 Industrial Ventilation 


: 
|| 
\ uy om SS 
ee | | maerents oe 
(AVA  ) 
WA ay LY 


FIGURE 1-5, Variation of SP, VP, and TP through a ventilation system 


SP, + VP, = SP, + VP, +h, [1.7] 
where: 
subscript 1= some upstream point 


subscript 2= some downstream point 


hi= all energy losses encountered by the air as 
it flows from the upstream to the down- 
stream point 


Note that, according to this principle, the total pressure 
must fall in the direction of flow. 


The application of these principles will be demonstrated by 


an analysis of the simple system shown in Figure 1-5. The 
normally vertical exhaust stack is shown laying horizontally 
to facilitate graphing the variation of static, total, and velocity 
pressures. The grinder wheel hood requires 300 cfm and the 
duct diameter is constant at 3.5 inches (0.0668 ft? area). 


1.6 ACCELERATION OF AIR AND HOOD ENTRY 
LOSSES 


Air flows from the room (point | of Figure 1-5) through 
the hood to the duct (point 2 of Figure 1-5) where the velocity 
can be calculated by the basic equation: 


_Q_ 300 


“A 0.0668 


= 4490 fom 


This velocity corresponds to a velocity pressure of 1.26 "wg, 
assuming standard air. 


If there are no losses associated with entry into a hood, then 
applying the energy conservation principle (Equation 1.7) to 
the hood yields 


SP, + VP, = SP, + VP, 


This is the well known Bernoulli principle of fluid mechan- 
ics. Subscript | refers to the room conditions where the static 
pressure is atmospheric (SP; = 0) and the air velocity is 
assumed to be very close to zero (VP, = 0). Therefore, the 
energy principle yields 


SP, = -VP, = -126 "wg 
Even if there were no losses, the static pressure must decrease 
due to the acceleration of air to the duct velocity. 


In reality, there are losses as the air enters the hood. These 
hood entry losses (hq) are normally expressed as a loss coef- 
ficient (Fy) multiplied by the duct velocity pressure; so hg = 
FyVPg (where VP = VP2). The energy conservation principle 
then becomes 


SP, = -(VP, +hy) [1.8] 
(See 3.5.1, 3.5.2, and Figure 5-1 for a discussion of h, and h,.) 


The absolute value of SP, is known as the hood static 
suction (SP,). Then 


SP, = -SP, = VP, +h, [1.9] 
For the example in Figure 1-5, assuming an entry loss coeffi- 
cient of 0.40, 

SP, = VP, + FyVP 
= 1.26 + (0.40) (1.26) 
= 1.26+ 0.50 = 1.76 "wg 

In summary, the static pressure downstream of the hood is 

negative (less than atmospheric) due to two effects: 
1. Acceleration of air to the duct velocity; and 
2. Hood entry losses. 


From the graph, note that TP, = — h,, which confirms the 
premise that total pressure decreases in the flow direction. 


An alternate method of describing hood entry losses is by 
the hood entry coefficient (C,). This coefficient is defined as 
the square root of the ratio of duct velocity pressure to hood 
static suction, or 


vP 
C, = Se [1.10] 


General Principles of Ventilation 1-7 


If there were no losses, then SP,,= VP and C, = 1.00. However, 
as hoods always have some losses, C, is always less than 1.00. 
In Figure 1-5, 


C.= vee 2 es = 0.845 
SP, 176 
An important feature of C, is that it is a constant for any given 


hood. It can, therefore, be used to determine the flow rate if 
the hood static suction is known. This is because 


Q = VA = 1096A es =1096A C, [SP [1.14] 
p p 


For standard air, this equation becomes 


Q = 4005 A C,,/SP, [1.12] 


For the example in Figure 1-5, 


Q = 4005(0.0668)(0.845) V176 = 300 cfm 


By use of C, and a measurement of SP), the flow rate of a 
hood can be quickly determined and corrective action can be 
taken ifthe calculated flow rate does not agree with the design 
flow rate. 


1.7 DUCT LOSSES 


There are two components to the overall total pressure 
losses in a duct run: 1) friction losses and 2) fitting losses. 


1.7.1 Friction Losses. Losses due to friction in ducts are 
a complicated function of duct velocity, duct diameter, air 
density, air viscosity, and duct surface roughness. The effects 
of velocity, diameter, density, and viscosity are combined into 
the Reynolds number (R.), as given by 


R, =—— [1.13] 


p= density, lbm/ft* 

d= diameter, ft 

v= velocity, ft/sec 

y= the air viscosity, Ibm/s-ft 


The effect of surface roughness is typically given by the 
relative roughness, which is the ratio of the absolute surface 
roughness height (k), defined as the average height of the 
roughness elements on a particular type of material, to the 
duct diameter. Some standard values of absolute surface 
roughness used in ventilation systems are given in Table 1-1. 


L. F. Moody“) combined these effects into a single chart 
commonly called the Moody diagram (see Figure 1-6). With 
a knowledge of both the Reynolds number and the relative 
roughness, the friction coefficient (f), can be found. 


1-8 Industrial Ventilation 


TABLE 1-1. Absolute Surface Roughness 


Duct Material Surface Roughness (k), feet 


Galvanized metal 0.00055 
Black iron 0.00015 
Aluminum 0.00015 
Stainless steel 0.00015 
Flexible duct 0.01005 
(wires exposed) 

Flexible duct 0.00301 
(wires covered) 


The above roughness heights are design values. It should be noted that 
significant variations from these values may occur, depending on the 
manufacturing process. 


Once determined, the friction coefficient is used in the 
Darcy—Weisbach friction coefficient equation to determine 
the overall duct friction losses: 


L 
het Ne [1.14] 


where: 
h, = friction losses in a duct, "wg 
f= Moody diagram friction coefficient (dimen- 


sionless) 
L= duct length, ft 
d= duct diameter, ft 
VP = duct velocity pressure, "wg 
There are many equations available for computer solutions 

to the Moody diagram. One of these is that of Churchill,@ 
which gives accurate (to within a few percent) results over the 
entire range of laminar, critical, and turbulent flow, all in a 
single equation. This equation is: 


12 4/12 
8 
f= (2) +(A +973] 
Re 


where: 


n-fewrl (ct) 


16 
B= ( 37,530 
Re 
While useful, this equation is quite difficult to use without 


a computer. Several attempts have been made to simplify the 
determination of friction losses for specialized situations. For 


[1.15] 


Q.4 “ye tamer ieat . ‘sient _ _— - na y wre Puree . 
oot =e TE EEE NIP Sor HE 
taal = AIEEE EEE HIEHTHES PHA 
we ait pain I. TURBULENCE, | 500 RIP 22 EE sail ; Hi 
on ae Hoe ee os 
Ht v COT 
co HALE SUE LIEH FEC) SER ET| SEY RENIN 
a APC sae ll UNEAEEUINIGH IEEE patti 
: 1% 43% a . res ae ee 
Se ORS TE PH 
04h ai) = rH ees Hr “EHH 
sig HE aay Re con hg wlo 
Sp. CE oan ht FATA Pt HEEEEEtH 
= aa ae ia s n 006 & 
OCH — eA LY PSE Eratmeeee ee 
eiser: SSaeiceeeemahiil Erect oo 
© 025 HHH K i =ae H +H 3 
Y ms cee ’ ——— 4 LH & 
ieee PSSST 
oan 7 saSNaN! : a 2 
o Ee Ee SN Ss ot = H SHH oo 
2) Ltt ft. 44dt NS ay es es Cor -0008 
& occ Cie : =a a Ee eet Soe, & 
oo HE FE TET Fob SES EEE 
HH LICL 1 PaaS CRS Eo 002 
rtf eer yet } ry = .0001 
ai ata SRE THOTT 
olf EL IEAIEET TABS et ome 
ooo HHT fy) ane : SSS Er 
ads Terr nia CCU bb “LLU PRS HH oo0.01 
t I 7 0 456 8 i 
fev) abo) 3 4 66 Bios 20) 3 456 B95 210) 3 4 £6 B IC6 OY 3S 456 Bip ee 
£. 
REYNOLDS NUMBER Re = uv 7 200.09, *000.09, 


FIGURE 1-6. Moody diagram (adapted from reference 1.1) 


many years, charts based on the Wright“) equation have been 
used in ventilation system design: 


(v/1000)'9 


hy = 274-5 


[1.16] 
where: 

V = duct velocity, fpm 

D = duct diameter, inches 


This equation gives the friction losses, expressed as "wg per 
100 feet of pipe, for standard air of 0.075 Ibm/ft’ density 
flowing through average, clean, round galvanized pipe having 
approximately 40 slip joints per 100 feet (k = 0.0005 ft). 


The later work by Loeffler'*) presented equations for use 
in the "velocity pressure" calculation method. Using the 
standard values of surface roughness, equations were ob- 
tained that could be used with the Darcy—Weisbach equation 
in the form: 


f 
hy = Ge VP =H,L VP u.t7 


where the "12" is used to convert the diameter D in inches to 
feet. 


Simplified equations were determined for the flow of 
standard air through various types of duct material with good 
accuracy (less than 5% error). The equations thus resulting 
were: 


[1.18] 


where the constant "a" and the exponents "b" and "c" vary 
as a function of the duct material as shown in Table 1-2. 
Note that no correlation was made with the extremely rough 
flexible duct with wires exposed. This equation, using the 
constants from Table 1-2 for galvanized sheet duct, were 
used to develop the friction Tables 5-5 and 5-6. Note that 
the value obtained from the chart or from equation 1.18 must 
be multiplied by both the length of duct and the velocity 
pressure. 


1.7.2 Fitting Losses. The fittings (elbows, entries, etc.) in 
a duct run will also produce a loss in total pressure. These 
losses are given in Chapter 5. 


The fitting losses are given by a loss coefficient (F) multi- 
plied by the duct velocity pressure. Thus, 


Ren = Fen VP [1.19] 


In contractions, entries, or expansions, there are several dif- 
ferent velocity pressures. The proper one to use with the loss 
coefficient will be identified where the coefficients are 
listed. 


In Figure 1-5, 15 feet of straight, constant diameter galva- 


General Principles of Ventilation 1-9 


TABLE 1-2. Correlation Equation Constants 
Duct Material k, Ft a b c 
0.00015 0.0425 0.465 0.602 


Aluminum, black iron, 
stainless steel 


Galvanized sheet duct 0.00051 0.0307 0.533 0.612 


Flexible duct, fabric 0.0035 0.0311 0.604 0.639 
wires covered 


nized duct connects the hood to a fan inlet. Because the duct 
area is constant, the velocity, and therefore the velocity pres- 
sure, is also constant for any given flow rate. The energy 
principle is: 


SP, + VP, = SP; + VP, +h, 


where subscript 3 refers to the fan inlet location. Because VP, 
= VP,, the losses will appear as a reduction in static pressure 
(there will, of course, be a corresponding reduction in total 
pressure). The friction loss can be found from Equation 1.17 
with the aid of Equation 1.18: 


0.533 


4490°-583 


= 0.0307 ——_ = 0.0828 
300 


0.612 
From Equation 1.17, h, = (0.0828)(15)(1.26) = 1.56 "wg. 
Using this in the energy principle, 


SP, = SP, —h, = -176 "wg-156 "wg = -3.32 "wg 


Another 10 feet of straight duct is connected to the dis- 
charge side of the fan. The losses from the fan to the end of 
the system would be about 1.04 "wg. Because the static 
pressure at the end of the duct must be atmospheric (SP; = 0), 
the energy principle results in 


SP, = SP; +h, =0 "wg+1.04 "wg = 104 "wg 


Therefore, the static pressure at the fan outlet must be 
higher than atmospheric by an amount equal to the losses in 
the discharge duct. 


1.8 MULTIPLE-HOOD EXHAUST SYSTEMS 


Most exhaust systems are more complicated than the pre- 
ceding example. It is usually more economical to purchase a 
single fan and air cleaner to service a series of similar opera- 
tions than to create a complete system for each operation. For 
example, the exhaust from 10 continuously used grinders can 
be combined into a single flow which leads to a common air 
cleaner and fan. This situation is handled similarly to a simple 
system, but with some provision to ensure that the air flow 
from each hood is as desired (see Chapter 5). 


1-10 Industrial Ventilation 


4000 FRM AIR 
VELOCITY AT 
FACE OF BOTH 


FIGURE 1-7. Blowing vs. exhausting 


1.9 AIR FLOW CHARACTERISTICS OF BLOWING AND 
EXHAUSTING 


Air blown from a small opening retains its directional effect 
for a considerable distance beyond the plane of the opening. 
However, if the flow of air through the same opening were 
reversed so that it operated as an exhaust opening handling 
the same volumetric flow rate, the flow would become almost 
non-directional and its range of influence would be greatly 
reduced. For this reason, local exhaust must not be contem- 
plated for any process that cannot be conducted in the imme- 
diate vicinity of the hood. Also, because of this effect, every 
effort should be made to enclose the operation as much as 
possible. Figure 1-7 illustrates the fundamental difference 
between blowing and exhausting. 


This effect also shows how the supply or replacement air 
discharge grilles can influence an exhaust system. If care is 


400 FPM 


EXHAUSTING 


APPROXIMATELY 10% OF FACE VELOCITY 
AT ONE DIA, AWAY FROM EXHAUST 
OPENING. 


not taken, the discharge pattern from a supply grille could 
seriously affect the flow pattern in front of an exhaust hood. 


REFERENCES 


1.1. Moody, L.F.: Friction Factors for Pipe Flow. ASME 
Trans. 66:672 (1944). 


1.2. Churchill, S.W.: Friction Factor Equation Spans All 
Fluid Flow Regimes. Chemical Engineering, Vol. 84 
(1977). 


1.3. Wright, Jr., D.K.: A new Friction Chart for Round 
Ducts. ASHVE Trans., Vol. 51, Appendix I, p. 312 
(1945). 


1.4. Loeffler, J.J.: Simplified Equations for HVAC Duct 
Friction Factors. ASHRAE J., p. 76 (January 1980). 


Chapter 2 
GENERAL INDUSTRIAL VENTILATION 


2.1 INTRODUCTION ............-....... 2-2 
2.2. DILUTION VENTILATION PRINCIPLES ...... 2-2 
2.3. DILUTION VENTILATION FOR HEALTH...... 2-2 

2.3.1. General Dilution Ventilation Equation ..... 2-2 

2.3.2 Calculating Dilution Ventilation for 

Steady State Concentration. .......... 2-5 

2.3.3 Contaminant Concentration Buildup ...... 2-5 

2.3.4 RateofPurging ................. 2-6 
2.4 MIXTURES—DILUTION VENTILATION 

FOR HEALTH.............-2-.-02--.- 2-6 
2.5. DILUTION VENTILATION FOR FIRE 

AND EXPLOSION ..............-.2004 2-7 
2.6 FIRE DILUTION VENTILATION FOR MIXTURES 2-8 
2.7. VENTILATION FOR HEAT CONTROL ....... 2-8 
2.8 HEAT BALANCE AND EXCHANGE......... 2-8 

2.8.1 SConvectiOn, 7. $k. bse eek wa geass 2-9 

2:68.20 “Radiaion: in34 6.8 hae Pha LS 2-9 

2.8.3. Evaporation... ............000- 2-9 
2.9 ADAPTIVE MECHANISM OF THE BODY...... 2-9 
2.10 ACCLIMATIZATION .............0.. 2-10 
Figure 2-1 "K" Factors Suggested for Inlet and Exhaust 

LOC ations ncn sae tk AE ea ie Me 2-4 

Figure 2-2 Contaminant Concentration Buildup... .. . . 2-6 
Figure 2-3 Rate of Purging .. 2.2... .....0200-. 2-6 
Figure 2-4 Heat Losses, Storage, and Temperature Relations 2-10 
Figure 2-5 Determination of Wet Bulb Globe Temperature 2-11 
Figure 2-6 Recommended Heat-Stress Alert Limits... . . 2-13 


2.11 ACUTE HEAT DISORDERS ............. 2-10 
24733) Heat Stroke: a0 she ote Sa ge otek BS 2-10 
2.11.2 Heat Exhaustion... ............. 2-10 
2.11.3 Heat Cramps and Heat Rash. ......... 2-11 

2.12 ASSESSMENT OF HEAT STRESS AND HEAT 

STRAIN, si.:¢ 'nhe-dosr-g ied Adie da Shab, weg tand ayeeds ath 2-H 
2.12.1 Evaluation of Heat Stress .. 0.0.0.2... 2-11 
2.12.2 Evaluationof Heat Strain .......... 2-12 

2.13 WORKER PROTECTION ............... 2-13 

2.14 VENTILATION CONTROL.............. 2-13 

2.15 VENTILATION SYSTEMS .............. 2-13 

2.16 VELOCITY COOLING ................ 2-15 

2.17 RADIANT HEAT CONTROL ............. 2-15 

2.18 PROTECTIVE SUITS FOR SHORT EXPOSURES — 2-16 

2.19 RESPIRATORY HEAT EXCHANGERS ....... 2-16 

2.20 REFRIGERATED SUITS ............... 2-16 

2.21 ENCLOSURES... .2...0..0.0-5 2020 n ae 2-16 

2:22: INSULATION ° <3 athegie geet enh 3h oe Bets 2-16 

REFERENCES 063 Ai See a eek teal 2-17 

Figure 2-7 Recommended Heat-Stress Exposure Limits . . 2-14 

Figure 2-8 Natural Ventilation ............... 2-14 

Figure 2-9 Mechanical Ventilation ............. 2-14 

Figure 2-10 Spot Cooling With Volume and Directional 

Control: \3. Je el acest tes: Be se els eS 2-16 

Figure 2-11 Heat Shielding... 2... 2.2... 2. 2-16 


2-2 Industrial Ventilation 


2.1 INTRODUCTION 


"General industrial ventilation" is a broad term which refers 
to the supply and exhaust of air with respect to an area, room, 
or building. It can be divided further into specific functions 
as follows: 


1. Dilution Ventilation is the dilution of contaminated air 
with uncontaminated air for the purpose of controlling 
potential airborne health hazards, fire and explosive 
conditions, odors, and nuisance-type contaminants. 
Dilution ventilation also can include the control of 
airborne contaminants (vapors, gases, and particu- 
lates) generated within tight buildings. 


Dilution ventilation is not as satisfactory for health 
hazard control as is local exhaust ventilation. Circum- 
stances may be found in which dilution ventilation 
provides an adequate amount of control more eco- 
nomically than a local exhaust system. One should be 
careful, however, not to base the economical consid- 
erations entirely upon the first cost of the system since 
dilution ventilation frequently exhausts large amounts 
of heat from a building, which may greatly increase 
the energy cost of the operation. 


2. Heat Control Ventilation is the control of indoor at- 
mospheric conditions associated with hot industrial 
environments such as are found in foundries, laun- 
dries, bakeries, etc., for the purpose of preventing 
acute discomfort or injury. 


2.2 DILUTION VENTILATION PRINCIPLES 


‘The principles of dilution ventilation system design are as 
follows: 


1. Select from available data the amount of air required 
for satisfactory dilution of the contaminant. The values 
tabulated on Table 2-1 assume perfect distribution and 
dilution of the air and solvent vapors. These values 
must be multiplied by the selected K value (see Section 
93:1). 


2. Locate the exhaust openings near the sources of con- 
tamination, if possible, in order to obtain the benefit 
of "spot ventilation." 


3. Locate the air supply and exhaust outlets such that the 
air passes through the zone of contamination. The 
operator should remain between the air supply and the 
source of the contaminant. 


4, Replace exhausted air by use of a replacement air 
system. This replacement air should be heated during 
cold weather. Dilution ventilation systems usually 
handle large quantities of air by means of low pressure 
fans. Replacement air must be provided if the system 
is to operate satisfactorily. 


5. Avoid re-entry of the exhausted air by discharging the 


exhaust high above the roof line or by assuring that no 
window, outdoor air intakes, or other such openings 
are located near the exhaust discharge. 


2.3 DILUTION VENTILATION FOR HEALTH 


The use of dilution ventilation for health has four limiting 
factors: 1) the quantity of contaminant generated must not be 
too great or the air flow rate necessary for dilution will be 
impractical; 2) workers must be far enough away from the 
contaminant source or the evolution of contaminant must be 
in sufficiently low concentrations so that workers will not 
have an exposure in excess of the established TLV; 3) the 
toxicity of the contaminant must be low; and 4) the evolution 
of contaminants must be reasonably uniform. 


Dilution ventilation is used most often to control the vapors 
from organic liquids with a TLV of 100 ppm or higher. In 
order to successfully apply the principles of dilution to such 
a problem, factual data are needed on the rate of vapor 
generation or on the rate of liquid evaporation. Usually such 
data can be obtained from the plant if any type of adequate 
records on material consumption are kept. 


2.3.1 General Dilution Ventilation Equation: The venti- 
lation rate needed to maintain a constant concentration at a 
uniform generation rate is derived by starting with a funda- 
mental material balance and assuming no contaminant in the 


air supply, 


Rate of Accumulation = Rate of Generation — 
Rate of Removal 


or 
VdC = Gdt- Q'Cadt [2.1] 
where: 
V = volume of room 
G= rate of generation 
Q’= effective volumetric flow rate 
C = concentration of gas or vapor 
t= time 
At a steady state, dc =0 


Gdt = Q’Cadt 


t2 t2 
I cat= | Q’Cadt 
4 4 


Ata constant concentration, C, and uniform generation rate, G, 
G(t, -t,) = Q’C (t, -t) 
Q’=— [2.2] 


Due to incomplete mixing, a K value is introduced to the rate 
of ventilation; thus: 


TABLE 2-1. Dilution Air Volumes for Vapors 


General Industrial Ventilation 2-3 


The following values are tabulated using the TLV values shown in parentheses, parts per million. TLV values are subject to revision if further research or 
experience indicates the need. If the TLV value has changed, the dilution air requirements must be recalculated. The values on the table must be 


multiplied by the evaporation rate (pts/min) to yield the effective ventilation rate (Q’) (see Equation 2.5). 
Ft? of Air (STP) Required for Dilution to TLV* 


Liquid (TLV in ppm)** 


Per Pint Evaporation 


Acetone (500) 

n-Amyl acetate (100) 

Benzene (0.5) 

n-Butanol (butyl alcohol) (50) 

n-Butyl acetate (150) 

Buty! Cellosolve (2-butoxyethanol) (25) 
Carbon disulfide (10) 

Carbon tetrachloride (5) 

Cellosolve (2-ethoxyethanol) (5) 
Cellosolve acetate (2-ethoxyethyl acetate) (5) 
Chloroform (10) 

1-2 Dichloroethane (ethylene dichloride) (10) 
1-2 Dichloroethylene (200) 

Dioxane (25) 

Ethyl acetate (400) 

Ethyl alcohol (1000) 

Ethyl ether (400) 

Gasoline (300) 

Isoamy! alcohol (100) 

Isopropyl alcohol (400) 

Isopropyl ether (250) 

Methyl acetate (200) 

Methyl alcohol (200) 

Methyl n-butyl ketone (5) 

Methyl Cellosolve (2-methoxyethanol) (5) 
Methyl Cellosoive acetate (2-methoxyethyl acetate) (5) 
Methyl chloroform (350) 

Methyl ethyl ketone (200) 

Methy! isobutyl ketone (50) 

Methyl propyl! ketone (200) 

Naphtha (coal tar) 

Naphtha VM&P (300) 

Nitrobenzene (1) 

n-Propyl acetate (200) 

Stoddard solvent (100) 
1,1,2,2-Tetrachloroethane (1) 
Tetrachloroethylene (25) 

Toluene (50) 

Trichloroethylene (50) 

Xylene (100) 


*The tabulated dilution air quantities must be multiplied by the selected K value. 


11,025 
27,200 
NOT RECOMMENDED 
88,000 
20,400 
NOT RECOMMENDED 
NOT RECOMMENDED 
NOT RECOMMENDED 
NOT RECOMMENDED 
NOT RECOMMENDED 
NOT RECOMMENDED 
NOT RECOMMENDED 
26,900 
NOT RECOMMENDED 
10,300 
6,900 
9,630 
REQUIRES SPECIAL CONSIDERATION 
37,200 
13,200 
11,400 
25,000 
49,100 
NOT RECOMMENDED 
NOT RECOMMENDED 
NOT RECOMMENDED 
11,390 
22,500 
64,600 
19,900 
REQUIRES SPECIAL CONSIDERATION 
REQUIRES SPECIAL CONSIDERATION 
NOT RECOMMENDED 
17,500 
30,000-35,000 
NOT RECOMMENDED 
159,400 
75,700 
90,000 


33,000 
“See Threshold Limit Values 1997 Appendix A. 


2-4 Industrial Ventilation 


a5 i Be BEST AIR INLET ne 
BES S? EXHAUST BEST EXHAUST 3EST EXHAUST 
K os. 1.0 MINIMUM K = 1.0 MINIMUM Ko o=1,.5 MINIMUM 


= “A \ 


PAIR AIR INL 
BEST EXHAUST 
= 2.5 MINIMUM 


REF. 2.2- 


a at Te it Ge ae 
we % \ a he 


K = 5 TO 10 
REF. 2.2 


NOTE: TRE K FACTORS LISTED HERE CONSIDER ONLY THE INLET AND EXHAUST 
AND ARE JUDGEMENTAL. TO SELECT THE K FACTOR USED IN THE EQUATION THE 
NUMBER AND LOCATION OF THE EMPLOYEES, THE SOURCE OF THE CONTAMINANT, 
AND THE TOXICITY OF THE CONTAMINANT MUST ALSO BE CONSIDERED. 


AMERICAN CONFERENCE | °K” FACTORS 
SUGGESTED FOR INLET 


OF GOVERNMENTAL AND EXHAUST LOCATIONS _ 


| INDUSTRIAL HYGIE NISTS ae 


Q’= = [2.3] 
where: 
Q = actual ventilation rate, cfm 
Q’ = effective ventilation rate, cfm 
K = a factor to allow for incomplete mixing 
Equation 2.2 then becomes: 
a= (2) K [2.4] 


This K factor is based on several considerations: 


1. The efficiency of mixing and distribution of replace- 
ment air introduced into the room or space being 
ventilated (see Figure 2-1). 


2. The toxicity of the solvent. Although TLV and toxicity 
are not synonymous, the following guidelines have 
been suggested for choosing the appropriate K value: 


Slightly toxic material: TLV > 500 ppm 


Moderately toxic material: TLV<S100-500 ppm 
Highly toxic material: TLV< 100 ppm 


3. A judgement of any other circumstances which the 
industrial hygienist determined to be of importance 
based on experience and the individual problem. In- 
cluded in these criteria are such considerations as: 


a. Duration of the process, operational cycle, and 
normal locations of workers relative to sources of 
contamination. 


b. Location and number of points of generation of the 
contaminant in the workroom or area. 


c. Seasonal changes in the amount ofnatural ventilation. 


d. Reduction in operational effectiveness of mechani- 
cal air moving devices. 


e. Other circumstances which may affect the concen- 
tration of hazardous material in the breathing zone 
of the workers. 


The K value selected, depending on the above considerations, 
ranges from | to 10. 


2.3.2 Calculating Dilution Ventilation for Steady-State 
Concentration: The concentration of a gas or vapor at a 
steady state can be expressed by the material balance equation 


ana 
Therefore, the rate of flow of uncontaminated air required to 
maintain the atmospheric concentration of a hazardous mate- 
rial at an acceptable level can be easily calculated if the 
generation rate can be determined. Usually, the acceptable 
concentration (C) expressed in parts per million (ppm) is 


General Industrial Ventilation 2-5 


considered to be the TLV. For liquid solvents, the rate of 
generation is 


G= CONSTANT x SGxER 
MW 
where: 
G= generation rate, cfm 


CONSTANT = _ the volume in ft* that | pt of liquid, when 
vaporized, will occupy at STP, ft*/pt 


SG = Specific gravity of volatile liquid 
ER = evaporation rate of liquid, pts/min 
MW = molecular weight of liquid 


Thus, Q’ = G + C can be expressed as 


gy - 403x10° x SGXER ia 
MW xC [2.5] 
EXAMPLE PROBLEM 


Methyl! chloroform is lost by evaporation from a tank at a 
rate of 1.5 pints per 60 minutes. What is the effective venti- 
lation rate (Q’) and the actual ventilation rate (Q) required to 
maintain the vapor concentration at the TLV? 


TLV = 350 ppm, SG = 132, MW = 133.4, Assume K=5 


Assuming perfect dilution, the effective ventilation rate (Q’) is 


Q’ 
(133.4) (350) 
For incomplete mixing, the actual ventilation rate (Q) is 


_ (403) (10°) (432) (15/60) (5) 
7 (133.4) (350) 


Q 


2.3.3 Contaminant Concentration Buildup (sce Figure 
2-2): The concentration of a contaminant can be calculated 
after any change of time. Rearranging the differential material 
balance results in 


dc dt 
G-QC V 


which can be integrated to yield 


nf Q’Co __ Q(tp- 4) a 


G-a'C, V 


where subscript 1 refers to the initial condition and subscript 
2 refers to the final condition. If it is desired to calculate the 
time required to reach a given concentration, then rearranging 
t, — t,, or At, gives 


Vv G-Q’C. 
At Ninh ee 
o { Ga, | [2.7] 


2-6 Industrial Ventilation 


STEADY STAT 


C ais eee ae asa eree 


TIME, " 
FIGURE 2-2. Contaminant concentration buildup 


If C, = 0, then the equation becomes 


__ Vi, (G-a'c, 
MSee Ge ) [2.8] 


Note: the concentration C, is ppm or parts/10° (e.g., if C, = 
200 ppm, enter C, as 200 + 10°). 


If it is desired to determine the concentration level (C2) 
after a certain time interval, t, — t, or At, and if C, = 0, then 
the equation becomes 


[2.9] 


Note: to convert C, to ppm, multiply the answer by 10°. 
EXAMPLE 


Methy! chloroform vapor is being generated under the 
following conditions: G= 1.2 cfm; Q’ = 2000 cfm; V = 100,000 
cu ft; C; = 0; K = 3. How long before the concentration (C2) 
reaches 200 ppm or 200 + 10°? 


Peed [n( $=") = 20.3min 


Q’ 


Using the same values as in the preceding example, what 
will be the concentration after 60 minutes? 


35 
Gli-e* * 
C, Sag RO = 419 ppm 


2.3.4 Rate of Purging (see Figure 2-3): Where a quantity 
of air is contaminated but where further contamination or 
generation has ceased, the rate of decrease of concentration 


over a period of time is as follows: 


VdC = -Q’Cadt 


or, 


[- wig] 
C,=Ce [2.10] 


EXAMPLE 


In the room of the example in Section 2.3.3, assume that 
ventilation continues at the same rate (Q’ = 2000 cfm) but that 
the contaminating process is interrupted. How much time is 
required to reduce the concentration from 100 (C,) to 25 (C>) 
ppm? 


t,-t, = -uf 2] = 69.3 min 


In the problem above, if the concentration (C;) at t) is 100 
ppm, what will concentration (C,) be after 60 minutes (At)? 


_ Qat 


C= eal * ) = 30.1 ppm 
2.4 MIXTURES—DILUTION VENTILATION FOR HEALTH 
In many cases, the evaporating liquid for which dilution 
ventilation rates are being designed will consist of a mixture 


of solvents. The common procedure used in such instances is 
as follows. 


C4 


TION 


i 
a 


CONCENTRA 


FIGURE 2-3. Rate of purging 


When two or more hazardous substances are present, their 
combined effect, rather than that of either individually, should 
be given primary consideration. /n the absence of information 
to the contrary, the effects of the different hazards should be 
considered as additive. That is, if the sum of the following 
fractions, 


Ci. om 

Tv, * TLV, 7 wv, [2.11] 
1 2 n 

exceeds unity, then the threshold limit of the mixture should 
be considered as being exceeded. "C" indicates the observed 
atmospheric concentration and "TLV" the corresponding 
threshold limit. In the absence of information to the contrary, 
the dilution ventilation therefore should be calculated on the 
basis that the effect of the different hazards is additive. The 
air quantity required to dilute each component of the mixture 
to the required safe concentration is calculated, and the sum 
of the air quantities is used as the required dilution ventilation 
for the mixture. 


Exceptions to the above rule may be made when there is 
good reason to believe that the chief effects of the different 
harmful substances are not additive but independent, as when 
purely local effects on different organs of the body are pro- 
duced by the various components of the mixture. In such 
cases, the threshold limit ordinarily is exceeded only when at 
least one member of the series itself has a value exceeding 


unity, €.g., 


C, Cy 
=~ or 
TLV, TLV; 


Therefore, where two or more hazardous substances are pre- 
sent and it is known that the effects of the different substances 
are not additive but act independently on the different organs 
of the body, the required dilution ventilation for each compo- 
nent of the mixture should be calculated and the highest cfm 
thus obtained used as the dilution ventilation rate. 


EXAMPLE PROBLEM 


A cleaning and gluing operation is being performed; 
methyl ethyl ketone (MEK) and toluene are both being re- 
leased. Both have narcotic properties and the effects are 
considered additive. Air samples disclose concentrations of 
150 ppm MEK and 50 ppm toluene. Using the equation given, 
the sum of the fractions [(150+200) + (S0+50) = 1.75] is 
greater than unity and the TLV of the mixture is exceeded. 
The volumetric flow rate at standard conditions required for 
dilution of the mixture to the TLV would be as follows: 


Assume 2 pints of each is being released each 60 
min. Select a K value of 4 for MEK and a K value of 
5 for toluene; sp gr for MEK = 0.805, for toluene = 
0.866; MW for MEK = 72.1, for toluene = 92.13. 


General Industrial Ventilation 2-7 


(403) (0.805)(10°) (4) (2/60) _ 090 of 


Q for MEK = 
72.1x 200 


(403) (0.866) (10°) (5) (2/60) 
92.13 x50 


Q for toluene = = 12,627 cfm 


Q for mixture = 3000 + 12,627 = 15,627 cfm 


2.5 DILUTION VENTILATION FOR FIRE AND 
EXPLOSION 


Another function of dilution ventilation is to reduce the 
concentration of vapors within an enclosure to below the 
lower explosive limit. It should be stressed that this concept 
is never applied in cases where workers are exposed to the 
vapor. In such instances, dilution rates for health hazard 
control are always applied. The reason for this will be appar- 
ent when comparing TLVs and lower explosive limits (LELs). 


The TLV of xylene is 100 ppm. The LEL of xylene is 1% 
or 10,000 ppm. An atmosphere of xylene safe-guarded against 
fire and explosion usually will be kept below 25% of the LEL 
or 2500 ppm. Exposure to such an atmosphere may cause 
severe illness or death. However, in baking and drying ovens, 
in enclosed air drying spaces, within ventilation ductwork, 
etc., dilution ventilation for fire and explosion is used to keep 
the vapor concentration to below the LEL. 


Equation 2.5 can be modified to yield air quantities to dilute 
below the LEL. By substituting LEL for TLV: 


_ (403)(sp_gr liquid)(100)(ER)( 


Q Lae Si) (for Standard Air) [212] 
(MW liquid)(LEL)(B) 


Note 1. Since LEL is expressed in % (parts per 100) rather 
than ppm (parts per million as for the TLV), the 
coefficient of 1,000,000 becomes 100. 


2. S- is a safety coefficient which depends on the 
percent of the LEL necessary for safe conditions. 
In most ovens and drying enclosures, it has been 
found desirable to maintain vapor concentrations 
at not more than 25% of the LEL at all times in all 
parts of the oven. In properly ventilated continuous 
ovens, a Sr coefficient of 4 (25% of the LEL) is 
used. In batch ovens, with good air distribution, the 
existence of peak drying rates requires an Sr coef- 
ficient of 10 or 12 to maintain safe concentrations 
at all times. In non-recirculating or improperly 
ventilated batch or continuous ovens, larger S¢ 
coefficients may be necessary. 


3. B is a constant which takes into account the fact 
that the lower explosive limit of a solvent vapor or 
air mixture decreases at elevated temperatures. B 
= | for temperatures up to 250 F; B = 0.7 for 
temperatures above 250 F. 


2-8 Industrial Ventilation 


EXAMPLE PROBLEM 


A batch of enamel-dipped shelves is baked in a recirculat- 
ing oven at 350 F for 60 minutes. Volatiles in the enamel 
applied to the shelves consist of two pints of xylene. What 
oven ventilation rate, in cfm, is required to dilute the xylene 
vapor concentration within the oven to a safe limit at all times? 


For xylene, the LEL = 1.0%; sp gr = 0.88; MW = 106; Sf 
= 10; B = 0.7. From Equation 2.12: 


_ (403)(0.88)(2 / 60)(100)(10) 
(106)(10)(0.7) 


=159cfm 


Q 


Since the above equation is at standard conditions, the air 
flow rate must be converted from 70 F to 350 F (operating 
conditions): 


Q, =(Qsqp) (Ratio of Absolute Temperature) 


(460 F +350 F) 
(460 F+70 F) 


Gs 159 2° 
530 


= (Qstp) 


= 243 cfm 
EXAMPLE PROBLEM 


In many circumstances, solvent evaporation rate is non- 
uniform due to the process temperature or the manner of 
solvent use. 


A 6 ft diameter muller is used for mixing resin sand on a 
10-minute cycle. Each batch consists of 400 pounds of sand, 
19 pounds of resin, and 8 pints of ethyl alcohol (the ethy] 
alcohol evaporates in the first two minutes). What ventilation 
rate is required? 


For ethyl alcohol, LEL = 3.28%; sp gr = 0.789; MW = 
46.07; Sf=4; B= 1 


q = (403)(0.789)(3/2)(100)(4) _ 3.367 cfm 


(46.07) (3.28) (1) 

Another source of data is the National Board of Fire Un- 
derwriters’ Pamphlet #86, Standard for Class A Ovens and 
Furnaces.°» This contains a more complete list of solvents 
and their properties. In addition, it lists and describes anumber 
of safeguards and interlocks which must always be considered 
in connection with fire dilution ventilation. See also Refer- 
ence 2.4. 


2.6 FIRE DILUTION VENTILATION FOR MIXTURES 


It is common practice to regard the entire mixture as 
consisting of the components requiring the highest amount of 
dilution per unit liquid volume and to calculate the required 
air quantity on that basis. [This component would be the one 


with the highest value for sp gr/(MW)(LEL).] 


2.7 VENTILATION FOR HEAT CONTROL. 


Ventilation for heat control in a hot industrial environment 
is a specific application of general industrial ventilation. The 
primary function of the ventilation system is to prevent the 
acute discomfort, heat-induced illness and possible injury of 
those working in or generally occupying a designated hot 
industrial environment. Heat-induced occupational illnesses, 
injuries, or reduced productivity may occur in situations 
where the total heat load may exceed the defenses of the body 
and result in a heat stress situation. It follows, therefore, that 
a heat control ventilation system or other engineering contro] 
method must follow a physiological evaluation in terms of 
potential heat stress for the occupant in the hot industrial 
environment. 


Due to the complexity of conducting a physiological evalu- 
ation, the criteria presented here are limited to general con- 
siderations. It is strongly recommended, however, that the 
NIOSH Publication No. 86-113, Criteria for a Recommended 
Standard, Occupational Exposure to Hot Environments,@ be 
reviewed thoroughly in the process of developing the heat 
control ventilation system. 


The development of a ventilation system for a hot industrial 
environment usually includes the control of the ventilation air 
flow rate, velocity, temperature, humidity, and air flow path 
through the space in question. This may require inclusion of 
certain phases of mechanical air-conditioning engineering 
design which is outside the scope of this manual. The neces- 
sary engineering design criteria that may be required are 
available in appropriate publications of the American Society 
of Heating, Refrigeration and Air-Conditioning Engineers 
(ASHRAE) handbook series. 


2.8 HEAT BALANCE AND EXCHANGE 


An essential requirement for continued normal body func- 
tion is that the deep body core temperature be maintained 
within the acceptable range of about 37 C (98.6 F)+ 1C 

(1.8 F). To achieve this, body temperature equilibrium re- 
quires a constant exchange of heat between the body and the 
environment. The rate and amount of the heat exchange are 
governed by the fundamental laws of thermodynamics of heat 
exchange between objects. The amount of heat that must be 
exchanged is a function of 1) the total heat produced by the 
body (metabolic heat), which may range from about | kilo- 
calorie (kcal) per kilogram (kg) of body weight per hour (1.16 
watts) at rest to 5 kcal/kg body weight/hour (7 watts) for 
moderately hard industrial work; and 2) the heat gained, if 
any, from the environment. The rate of heat exchange with 
the environment is a function of air temperature and hu- 
midity; skin temperature; air velocity; evaporation of 
sweat; radiant temperature; and type, amount, and charac- 
teristics of the clothing worn, among other factors. Respi- 


ratory heat loss is of little consequence in human defenses 
against heatstress. 


The basic heat balance equation is: 
AS = (M—-W)+C+R-E [2.13] 


where: 
AS = change in body heat content 


(M — W) = total metabolism — external work per- 
formed 


= convective heat exchange 
= radiative heat exchange 
= evaporative heat loss 


To solve the equation, measurement of metabolic heat 
production, air temperature, air water vapor pressure, wind 
velocity, and mean radiant temperature are required. 


The major modes of heat exchange between man and the 
environment are convection, radiation, and evaporation. 
Other than for brief periods of body contact with hot tools, 
equipment, floors, etc., which may cause burns, conduction 
plays a minor role in industrial heat stress. Because of the 
typically small areas of contact between either body surfaces 
or clothing and hot or cold objects, heat exchange by thermal 
conduction is usually not evaluated in a heat balance equation 
for humans. The effect of heat exchange by thermal conduc- 
tion in human thermal regulation is important when large 
areas of the body are in contact with surfaces that are at 
temperatures different from average skin temperature (nomi- 
nally 95 F), e.g., when someone is prone or supine for long 
periods. It is also important when even small body areas are 
in contact with objects that provide steep thermal gradients 
for heat transfer, e.g., when someone is standing on very cold 
or very hot surfaces. 


The equations for calculating heat exchange by convection, 
radiation, and evaporation are available in Standard Interna- 
tional (SI) units, metric units, and English units. In SI units 
heat exchange is in watts per square meter of body surface 
(W/m?). The heat exchange equations are available in both 
metric and English units for both the seminude individual and 
the worker wearing conventional long-sleeved work shirt and 
trousers. The values are in kcal/h or British thermal units per 
hour (Btu/h) for the "standard worker" defined as one who 
weighs 70 kg (154 Ibs) and has a body surface area of 1.8 m? 
(19.4 ft?). 


2.8.1 Convection: The rate of convective heat exchange 
between the skin of a person and the ambient air immediately 
surrounding the skin is a function of the difference in tem- 
perature between the ambient air (t,), the mean weighted skin 
temperature (t,.) and the rate of air movement over the skin 
(V,). This relationship is stated algebraically for the "standard 
worker" wearing the customary one-layer work clothing en- 
semble as: 


General Industrial Ventilation 2-9 


C= 0.65 Vy (t, — te) [2.14] 


where: 
C = convective heat exchange, Btu/h 
V, = air velocity, fpm 
t, = air temperature, F 


ty, = mean weighted skin temperature, usually as- 
sumed to be 95 F 


When t, > 95 F there will be a gain in body heat from the 
ambient air by convection. When t, < 95 F, heat will be lost 
from the body to the ambient air by convection. 


2.8.2 Radiation: Infrared radiative heat exchange between 
the exposed surfaces of a person’s skin and clothing varies as 
a function of the difference between the fourth power of the 
absolute temperature of the exposed surfaces and that of the 
surface of the radiant source or sink, the exposed areas and 
their emissivities. Heat is gained by thermal radiation if the 
facing surface is warmer than the average temperature of the 
exposed skin and clothing, and vice versa. A practical ap- 
proximation for infrared radiant heat exchange for a person 
wearing conventional clothing is: 


R = 15.0 (tw-t,) [2.15] 


where: 


R = radiant heat exchange, Btu/h 


ty = mean radiant temperature, F 


t, = mean weighted skin temperature 


2.8.3 Evaporation: The evaporation of water (sweat) or 
other liquids from the skin or clothing surfaces results in a 
heat loss from the body. Evaporative heat loss for humans is 
a function of air flow over the skin and clothing surfaces, the 
water vapor partial pressure gradient between the skin surface 
and the surrounding air, the area from which water or other 
liquids are evaporating and mass transfer coefficients at their 
surfaces. 


E = 24V!°(p.¢ —Pa) [2.16] 


where: 
E = evaporative heat loss, Btu/h 


< 
ry 
| 


= air velocity, fpm 
P= water vapor pressure of ambient air, mmHg 


= water vapor pressure on the skin, assumed to be 
42 mm Hg at a 95 F skin temperature 


Bo) 

g 

x 
| 


2.9 ADAPTIVE MECHANISM OF THE BODY 


Even people in generally good health can adjust physiologi- 
cally to thermal stress only over a narrow range of environ- 
mental conditions. Unrestricted blood flow to the skin, an 
unimpeded flow of dry, cool air over the skin surface and 
sweating are prime defenses in heat stress. Although heat 


2-10 Industrial Ventilation 


HEAT LOSSES, STORAGE, AND TEMPERATURE 
RELATIONS FOR CLOTHED SUBJECT 


500 


Z 


Se 


eS () ec cate et Po Ne ee a ee ea 
60 7) 80 90 100 110 


DRY BULB TEMPERATURE, DEG. F 


FIGURE 2-4. Heat losses, storage, and temperature relations 


produced by muscle activity reduces the impact of cold stress, 
it can add substantially to the total challenge during heat 
stress. Diminished health status, medications, limited prior 
thermal exposure, among other factors, increase danger from 
thermal stresses. 


The reflex contro] of blood flow is the body’s most effec- 
tive and important first line of defense in facing either cold or 
heat stress. Reducing blood flow to the skin of the hands, feet, 
fingers and toes is an important measure for reducing heat loss 
in a cold environment. Blood flow to the skin, however, 
increases many-fold during heat stress. Its effect is to increase 
rates of heat distribution in the body and maximize conduc- 
tive, convective, radiative and evaporative heat losses to the 
environment (Figure 2-4). Its cost is often to reduce perfusion 
of other organs, especially the brain, and reduce systemic 
arterial blood pressure, leading to reduced consciousness, 
collapse, heat exhaustion and other heat-induced illnesses. 


Reflex sweating during the physical activities of exercise, 
work and/or heat stress brings often large volumes of body 
water and electrolytes (salts) to the skin surface. Heat is lost 
when the water in sweat evaporates. Whether the electrolytes 
remain on the skin surface or are deposited in clothing, they 
are nonetheless permanently lost to the body. The electrolyte 
content of a typical American diet usually provides adequate 
electrolyte replacement for these losses. Electrolyte replace- 
ment fluids, however, may be necessary for people on salt-re- 
stricted diets and those who commonly sustain periods of 
prolonged and profuse sweating. It is essential for everyone 
that the lost body water and electrolytes are replaced in the 
same volume and proportion as lost in sweat. Muscle spasms, 
cramps, gastrointestinal disturbances and general malaise, 


among other signs and symptoms, commonly develop when 
they are not. 


2.10 ACCLIMATIZATION 


People in generally good health normally develop heat 
acclimatization in a week or so after intermittently working 
or exercising in high heat. Its effect is to improve the comfort 
and safety of the heat exposure. It occurs because of an 
increase in total circulating blood volume, an improved ability 
to maintain systemic arterial blood pressure during heat stress, 
and a developed ability to produce larger volumes of more 
dilute sweat at rates of production more precisely matched to 
the heat load. Heat acclimatization rapidly diminishes even 
after a day or so of discontinued activity in the heat. Most is 
lost after about a week. 


2.11 ACUTE HEAT DISORDERS 


A variety of heat disorders can be distinguished clinically 
when individuals are exposed to excessive heat. A brief 
description of these disorders follows. 


2.11.1 Heat Stroke: Heat stroke (also known as "sun 
stroke") is a life-threatening condition which without excep- 
tion demands immediate emergency medical care and hospi- 
talization. Before medical care arrives, move the person to a 
shaded area, check for other injuries, ensure there is an 
unobstructed airway, remove or loosen clothing, and flood the 
body surface with free-flowing, tepid (not cold) water. Vig- 
orous fanning helps cooling. Heat stroke develops when body 
heat gains from exercise, work and/or a hot environment 
overwhelm normal thermoregulatory defenses. Charac- 
teristically, sweating has ceased, the skin is hot and dry, and 
deep body temperature is above about 104 F. The person may 
be either diaphoretic, semiconscious, unconscious or agitated, 
delirious and in convulsions. Demand medical care even if 
consciousness returns—lethal effects may develop in the next 
24 to 72 hours. 


2.11.2 Heat Exhaustion: Heat exhaustion (also called 
"exercise-induced heat exhaustion" and "heat syncope") most 
commonly occurs in people who are not heat acclimatized and 
who are in poor physical condition, obese, inappropriately 
dressed, and exercising or working energetically in the heat 
at unaccustomed and/or demanding tasks. It is characterized 
by lightheadedness, dizziness, vision disturbances, nausea, 
vague flu-like symptoms, tinnitus, weakness, and occasion- 
ally, collapse. The person’s deep body temperature is typi- 
cally in a normal range or only slightly elevated; the skin is 
moist and cool but may be reddened by its high rate of blood 
flow. Heat exhaustion develops when there is reflex demand 
for blood flow to the skin to dissipate body heat and a 
simultaneous reflex demand for blood flow to exercising 
muscles to meet metabolic needs of increased activity. These 
peripheral distributions of blood volume reduce systemic 
arterial pressure and brain blood flow, causing most of the 


TABLE 2-2. Estimating Energy Cost of Work by Task Analysis?) 


A. Body position and movement kcal/min* 
Sitting 0.3 
Standing 0.6 
Walking 2.0-3.0 
Walking uphill Add 0.8/meter rise 

Average Range 

B. Type of Work kcal/min kcal/min 
Hand work — light 0.4 0.2-1.2 
Hand work — heavy 0.9 
Work one arm — light 1.0 0.7-2.5 
Work one arm — heavy 1.7 
Work both arms - light 15 1.0-3.5 
Work both arms — heavy 25 
Work whole body — light 3.5 2.5-15.0 
Work whole body — moderate 5.0 
Work whole body — heavy 70 
Work whole body — very heavy 9.0 

C. Basal metabolism 1.0 

D. Sample calculation ** 

Assembling work with heavy hand tools 
1. Standing 0.6 
2. Two-arm work 3.5 
3, Basal metabolism 1.0 
TOTAL 5.1 kcal/min 


*For standard worker of 70 kg body weight (154 Ibs) and 1.8 m body surface 
(19.4 ft’). 


“Example of measuring metabolic heat production of a worker when performing 
initial screening. 


symptoms of heat exhaustion. Resting in a cool environment 
where there is free flowing, dry air usually remediates symp- 
toms quickly. Although heat exhaustion is debilitating and 
uncomfortable, it is not often a long-term health threat. There 
are considerable dangers, of course, for anyone operating 
machinery when consciousness is impaired because of heat 
exhaustion or for any other reason. 


2.11.3 Heat Cramps and Heat Rash: Heat cramps (also 
known as "muscle cramps") are spontaneous, involuntary, 
painful and prolonged muscle contractions that commonly 
occur in otherwise healthy people when both body water and 
electrolyte levels have not been restored after extended peri- 
ods of heavy sweating during exercise and/or heat stress. Full 
recovery can be expected in about 24 hours with the use of 
electrolyte replacement fluids and rest. Heat rash (also known 
as "prickly heat" or "miliaria rubia") is an acute, inflammatory 
skin disease characterized by small red, itchy or tingling 
lesions, commonly in areas of skin folds or where there is 


General Industrial Ventilation 2-11 


abrasive clothing. It commonly disappears when these areas 
are kept dry, unabraded and open to free flowing, dry air. 


2.12 ASSESSMENT OF HEAT STRESS AND HEAT 
STRAIN 


Heat Stress is defined by environmental measurements of 
air temperature, humidity, air flow rate, the level of radiant 
heat exchange and evaluation of a person’s metabolic heat 
production rate from exercise and/or work. Heat stress is the 
load on thermoregulation. Heat Strain is defined as the cost 
to each person facing heat stress. Although all people working 
at the same intensity in the same environment face the same 
level of heat stress, each is under a unique level of heat strain. 
Almost any environmental thermal exposure will be com fort- 
able and safe for some, but endangering, even lethal to others. 
Because disabilities, danger and death arise directly from heat 
strain, no measure of heat stress is a reliable indicator of a 
particular person’s heat strain or the safety of the exposure. 


2.12.1 Evaluation of Heat Stress: Dry-bulb air tempera- 
ture (DB: so-called "dry-bulb" temperature) is measured by 
calibrated thermometers, thermistors, thermocouples and 
similar temperature-sensing devices which themselves do not 
produce heat and which are protected from the effects of 
thermal conduction, evaporation, condensation and radiant 
heat sources and sinks. Relative humidity is evaluated psy- 
chrometrically as a function of the steady-state difference 
between dry-bulb temperature and that indicated by the tem- 
perature of a sensor covered with a freely evaporating, water- 
saturated cotton wick. Such a measure reports "NWB" 


D. B. THERMOMETER 
(USED ONLY OUTDOOR 
IN SUNSHINE) 
NATURAL W. B. 
THERMOMETER | 
GLOBE 
THERMOMETER 


Sin 
a 1 in || 
WICK 
=e —_ H) 
1 in I \ if 


125 ml FLASK WITH : 
DISTILLED WATER 6" COPPER SHELL 


PAINTED MATTE BLACK 
wick / 


_=E==)p 


TTT) 


FIGURE 2-5. Determination of wet-bulb globe temperature 


2-12 Industrial Ventilation 


(natural wet-bulb temperature) when the wetted sensor is 
affected only by prevailing air movement, and "WB" (when 
it is exposed to forced convection). Free air movement is 
measured with an unobstructed anemometer. Infrared radiant 
"heat transfer" is typically measured by a temperature sensor 
at the center of a 6-inch, hollow, copper sphere painted flat 
("matte") black. Such a measure reports "GT" (globe tempera- 
ture) (Figure 2-5). A person’s metabolic heat production is 
usually evaluated from an estimated level of average physical 
activity (Table 2-2). 


Although there are a number of different indices for evalu- 
ating heat stress, none is reliable as a sole indicator of heat 
strain for a specific person. Dry-bulb temperature is the least 
valuable measure of heat stress because it provides no infor- 
mation about ambient relative humidity, or heat exchange by 
convection or radiation, and gives no estimate of the metabo- 
lic heat production. Wet-bulb, globe temperature (WBGT) is 
often used as an index of heat stress. When there is a source 
of radiant heat transfer (solar radiation, hot surfaces of ma- 
chinery): 


WBGT = 0.7 tayp +0.2ty +0. t, [2.17] 


where 
tawb = natural wet-bulb temperature 
ty = globe temperature 


When radiant heat transfer is negligible: 
WBGT = 07 tyyp +0.3 ty [2.18] 


WBGT evaluates more factors contributing to heat stress than 
does dry-bulb temperature alone. It does not, however, effec- 
tively evaluate the importance of energy transfer from human 
skin by convection which is essential for the removal of heat 
from the skin surface and the formation of water vapor from 
secreted sweat. Nor does WBGT evaluate the importance of 
metabolic heat production in the heat stress. Under some 
environmental conditions, heat produced by metabolism is the 
predominant stressor. 


2.12.2 Evaluation of Heat Strain: The incidence and se- 
verity of heat strain will vary greatly among people exposed 
to the same level of heat stress. Paying attention to the early 
signs and symptoms of heat strain is the best first line of 
defense against debilitating heat-induced discomfort and in- 
juries. It is dangerous, inappropriate and irresponsible to 
consider a heat stress as safe for all when some exposed to it 
show heat strain signs and symptoms, while others do not. 
Acute heat strain is indicated by: 


Visible Sweating: Thermoregulatory reflexes nor- 
mally fine-tune with precision the rate of sweating to 
the rate at which body heat must be lost to maintain 
homeostasis. Normally, there is no liquid water on the 
skin surface in a tolerable heat stress because water 
brought to the skin surface by sweating readily forms 


invisible water vapor in the process of evaporative 
cooling. Although an all too common occurrence in 
the workplace, liquid sweat either on the skin surface, 
or soaked into clothing, is a sure sign of heat strain. It 
indicates the level of sweating required to keep body 
temperature in a normal range cannot be matched by 
the rate of water evaporation from the skin surface to 
the environment. It is necessary either to increase the 
air flow rate over skin and clothing surfaces, lower 
ambient temperature and relative humidity, reduce 
radiative heat gain, and/or reduce metabolic heat pro- 
duction if progressive heat disabilities are to be 
avoided. 


Discontinued Sweating: A hot, dry skin for someone 
exposed to heat stress is a dangerous sign. It indicates 
suppression of sweating, perhaps exacerbated by pre- 
scription or over-the-counter medications. The appear- 
ance of a hot, dry skin for someone in a heat stress 
demands immediate attention and corrective actions. 


Elevated Heart Rate: Short-term increases in heart 
rate are normal for episodic increases in work load. In 
a heat stress, however, a sustained heart rate greater 
than 160/min for those younger than about 35 years, 
or 140/min for those who are older, is a sign of heat 
strain. 


Elevated Deep Body Temperature: A sustained deep 
body temperature greater than 100.4 F is a sign of heat 
strain in someone exposed to heat stress. 


Decreased Systemic Arterial Blood Pressure: A fall 
in blood pressure of more than about 40 Torr in about 
3.5 minutes for someone working in a heat stress 
indicates a heat-induced disability. Reduced con- 
sciousness, feeling of weakness, vision disturbances, 
and other signs and symptoms are likely to follow. 


Personal Discomfort: Heat strain may be indicated in 
some heat-stressed individuals by severe and sudden 
fatigue, nausea, dizziness, lightheadedness, or faint- 
ing. Others may complain of irritability; mental con- 
fusion; clumsiness; forgetfulness; general malaise; the 
development of sometimes vague, flu-like symptoms; 
and paradoxical chills and shivering. 


Infrequent Urination: Urinating less frequently than 
normal and the voiding of a small volume of dark-col- 
ored urine is a sign of whole body dehydration. Dehy- 
dration compromises the body’s ability to maintain a 
large enough circulating blood volume so that normal 
blood pressure is maintained in the face of the com- 
bined stressors of exercise and heat exposure. People 
who work or exercise in the heat need to develop the 
habit of drinking adequate volumes of water at fre- 
quent enough intervals to maintain the same patterns 
of urination they have when not heat stressed. Those 
who sweat heavily for long periods need also to discuss 


with their physicians a possible need for using electro- 
lyte replacement fluids. 


2.13 WORKER PROTECTION 


There is improved safety, comfort and productivity when 
those working in the heat are: 


1. In generally good physical condition and not obese, 
are heat acclimatized, and are experienced in the heat 
stressing job. They also need to know how to select 
clothing and maintain whole body hydration and elec- 
trolyte levels to provide the greatest comfort and 
safety. 


2. In areas that are well-ventilated and shielded from 
infrared radiant heat sources. 


3. Knowledgeable about the effects of their medications 
on cardiovascular and peripheral vascular function, 
blood pressure control, body temperature mainte- 
nance, sweat gland activity, metabolic effects and 
levels of attention or consciousness. 


4. Appropriately supervised when there is a history of 
abuse or recovery from abuse of alcohol or other 
intoxicants. 


5. Provided accurate verbal and written instructions, fre- 
quent training programs and other information about 
heat stress and strain. 


| 
ihe 
be i 0414.0 
OO; sae 
| “ 
pie. ele. Sa, 
Iya | c 
Pech: 
wi /95)35 
a 
i o 
SS. Pe 
ae 
ae _ 
~e 
~~ 
76 SSS 
cag 


~ 300 


General Industrial Ventilation 2-13 


6. Able to recognize the signs and symptoms of heat 
strain in themselves and others exposed to heat stress 
and know the appropriately effective steps for their 
remediation (Figures 2-6 and 2-7). 


2.14 VENTILATION CONTROL 


The contro! method presented here is limited to a general 
engineering approach. Due to the complexity of evaluating a 
potential heat stress-producing situation, it is essential that the 
accepted industrial hygiene method of recognition, evalu- 
ation, and control be utilized to its fullest extent. In addition 
to the usual time-limited exposures, it may be necessary to 
specify additional protection which may include insulation, 
baffles, shields, partitions, personal protective equipment, 
administrative control, and other measures to prevent possible 
heat stress. Ventilation control measures may require a source 
of cooler replacement air, an evaporative or mechanically 
cooled source, a velocity cooling method, or any combination 
thereof. Specific guidelines, texts, and other publications or 
sources should be reviewed for the necessary data to develop 
the ventilation system. 


2.15 VENTILATION SYSTEMS 


Exhaust ventilation can be used to remove excessive heat 
and/or humidity if a replacement source of cooler air is 
available. If it is possible to enclose the heat source, such as 
is the case with ovens or certain furnaces, a gravity or forced 


eet 8 min. /h. \ 
= 30 min. /h.f 
ce ° RAL 
XY 04.5 min. /h. 


i, \y 
60 min. /h.) 


T2060 
% 


-OMMENDED ALERT LIMIT 


*FOR "STANDARD WORKER” OF 70 kg (154 tbs) BODY WEIGHT AND 
1.8 m2 (9.4 2) BODY SURFACE. 


FIGURE 2-6. Recommended heat-stress alert limits, heat-unacclimatized workers 


2-14 Industrial Ventilation 


a 


pes : 
cn|04 40 


= cee : min. /p.) 
e — 30 min. Jn Jd REI 
oe ON ae 45 min. /h.f ~ 
> i 3 AS in. /h.{ 
Zlyalos “6 0 min./h.) 
168}20L_ ag 

T0300 300 405 500 kcal /h. 

400 800 1200 1600 2000 Btu /h. 

116 233 349 465 289 wotts 


METABOLIC HEAT 


C = CEILING LIMIT 

REL = RECOMMENDED EXPOSURE LIMIT 

sFOR "STANDARD WORKER” OF 70 kg (154 ibs) BODY WEIGHT AND 
1.8 m2 (49.4 ft? ) BODY SURFACE. 


FIGURE 2-7. Recommended heat-stress exposure limits, heat-acclimatized workers 


air stack may be all that is necessary to remove excessive heat The sensible heat rise can be determined by the following: 
from the workroom. If a partial enclosure or local hood is 
indicated, control velocities should be used as described in H, = Q, xp xc, x AT x(60 min/hr) [2.19] 
Chapter 3. 

where: 


Many operations do not lend themselves to local exhaust. 


General ventilation may be the only alternative. To determine Hs = Sensible neat gain, Brune 

the required general ventilation, the designer must estimate Q, = Volumetric flow for sensible heat, cfm 
the acceptable temperature or humidity rise. The first step in p = Density of the air, lbm/ft® 

determining the required volumetric flow is to determine the cy = Specific heat of the air, BTU/Ibm- F 


sensible and latent heat load. Next, determine the volumetric 
flow to dissipate the sensible heat and the volumetric flow to 
dissipate the latent heat. The required general ventilation is For air c, = 0.24 BTU/Ibm ~— F and p = 0.075 Ibm/ft; 
the larger of the two volumetric flows. 


AT = Change in temperature, F 


paeee) ~ Se eee = ee 
a sh _ 

See 600 FPM 
TARGET VEL. 


FIGURE 2-8. Good natural ventilation and circulation FIGURE 2-9. Good mechanically supplied ventilation 


consequently, the equation becomes 
H, =108xQ, x AT 
or 
Q, =H, + (1.08 x AT) [2.20] 


In order to use this equation, it is necessary to first estimate 
the heat load. This will include loads from the sun, people, 
lights, and motors, as well as other particular sources of heat. 
Of these, sun load, lights, and motors are all completely 
sensible. The people heat load is part sensible and part latent. 
In the case of hot processes which give off both sensible and 
latent heat, it will be necessary to estimate the amounts or 
percents of each. In using the above equation for sensible heat, 
one must decide the amount of temperature rise which will be 
permitted. Thus, in a locality where 90 F outdoor dry bulb 
may be expected, if it is desired that the inside temperature 
not exceed 100 F, or a 10-degree rise, a certain air flow rate 
will be necessary. If an inside temperature of 95 F is required, 
the air flow rate will be doubled. 


For latent heat load, the procedure is similar, although more 
difficult. If the total amount of water vapor is known, the heat 
load can be estimated from the latent heat of vaporization, 970 
BTU/b. In a manner similar to the sensible heat calculations, 
the latent heat gain can be approximated by: 


H, = Q, xp xc, x Ah x (60 min/ hr) x (11!b/ 7000 grains) 


where: 
H, = Latent heat gain, BTU/hr 
Q, = Volumetric flow for latent heat, cfm 
p = Density of the air, lbm/ft? 
c, = Latent heat of vaporization, BTU/Ibm 


Ah = Change in absolute humidity of the air, grains- 
water/Ibm-dry air 


For air, ¢ is approximately 970 BTU/Ib and p = 0.075 
lbm/ft?. Consequently, the equation becomes 


H, =0.62xQ,x Ah 
or 


Hy 


(oes 
10.62 Ah [2.21] 


If the rate of moisture released, M in pounds per hours, is 
known, then 


M = Q4 xp x Ahx (1b / 7000 gr) x (60 min/ hr) 


=Q,xpxAh-= (116.7) 


or 


General Industrial Ventilation 2-15 


6.2 116.7xM 
arr a [2.22] 


The value of the "grains-water per pound-air difference" is 
read from a psychrometric chart or table. It represents the 
difference in moisture content of the outdoor air and the 
conditions acceptable to the engineer designing the exhaust 
system. The air quantities calculated from the two equations 
above should not be added to arrive at the required quantity. 
Rather, the higher quantity should be used since both sensible 
and latent heat are absorbed simultaneously. Furthermore, in 
the majority of cases the sensible heat load far exceeds the 
latent heat load, so the design usually can be calculated on the 
basis of sensible heat alone. 


The ventilation should be designed to flow through the hot 
environment in a manner that will efficiently control the 
excess heat. Figures 2-8 and 2-9 illustrate this principle. 


2.16 VELOCITY COOLING 


If the air dry-bulb or wet-bulb temperatures are lower than 
95-100 F, the worker may be cooled by convection or evapo- 
ration. When the dry bulb temperature is higher than 95-100 
F, increased air velocity may add heat to the worker by 
convection. If the wet bulb temperature is high also, evapora- 
tive heat loss may not increase proportionately and the net 
result will be an increase in the worker’s heat burden. Many 
designers consider that supply air temperature should not 
exceed 80 F for practical heat relief. 


Current practice indicates that air velocities in Table 2-3 
can be used successfully for direct cooling of workers. For 
best results, provide directional control of the air supply 
(Figure 2-10) to accommodate daily and seasonal variations 
in heat exposure and supply air temperature. 


2.17 RADIANT HEAT CONTROL 


Since radiant heat is a form of heat energy which needs no 
medium for its transfer, radiant heat cannot be controlled by 


TABLE 2-3. Acceptable Comfort Air Motion at the Worker 
Air Velocity, fpm* 


Continuous Exposure 
Air conditioned space 50-75 


Fixed work station, general ventilation 


or spot cooling: — Sitting 75-125 
Standing 100-200 
Intermittent Exposure, Spot Cooling or Relief Stations 
Light heat loads and activity 1000-2000 
Moderate heat loads and activity 2000-3000 
High heat loads and activity 3000-4000 


*Note: Velocities greater than 1000 fpm may seriously disrupt the performance of 
nearby local exhaust systems. Care must be taken to direct air motion to 
prevent such interference. 


2-16 Industrial Ventilation 


300 TO 3 
1000 TO 200¢ 


| INGOT LOADER 
FIGURE 2-10. Spot cooling with volume and directional control 


ventilation. Painting or coating the surface of hot bodies with 
materials having low radiation emission characteristics is one 
method of reducing radiation. 


For materials such as molten masses of metal or glass which 
cannot be controlled directly, radiation shields are effective. 
These shields can consist of metal plates, screens, or other 
material interposed between the source of radiant heat and the 
workers. Shielding reduces the radiant heat load by reflecting 
the major portion of the incident radiant heat away from the 
operator and by re-emitting to the operator only a portion of 
that radiant heat which has been absorbed. Table 2-4 indicates 
the percent of both reflection and emission of radiant heat 
associated with some common shielding materials. Addi- 
tional ventilation will control the sensible heat load but will 
have only a minimal effect, if any, upon the radiant heat load. 
See Figure 2-11. 


2.18 PROTECTIVE SUITS FOR SHORT EXPOSURES 


For brief exposures to very high temperatures, insulated 
aluminized suits and other protective clothing may be worn. 
These suits reduce the rate of heat gain by the body but provide 
no means of removing body heat; therefore, only short expo- 
sures may be tolerated. 


REFLEC TIVE 


SHIELD i 
~ 320 F TEU 


ae - 
i OOOOOoag ) 


& 
t 


as Pa oe 


FIGURE 2-11. Heat Shielding 


TABLE 2-4. Relative Efficiencies of Common Shielding Materials 
Reflection of 


Radiant Heat Emission of 

Incident Upon Radiant Heat 
Surface of Shielding Surface from Surface 
Aluminum, bright 95 5 
Zinc, bright 90 10 
Aluminum, oxidized 84 16 
Zinc, oxidized 73 27 
Aluminum paint, new, clean 65 35 
Aluminum paint, dull, dirty 40 60 
Iron, sheet, smooth 45 55 
iron, sheet, oxidized 35 65 
Brick 20 80 
Lacquer, black 10 90 
Lacquer, white 10 90 
Asbestos board 6 94 
Lacquer, flat black 3 97 


2.19 RESPIRATORY HEAT EXCHANGERS 


For brief exposure to air of good quality but high tempera- 
ture, a heat exchanger on a half-mask respirator face piece is 
available. This device will bring air into the respiratory pas- 
sages at a tolerable temperature but will not remove contami- 
nants nor furnish oxygen in poor atmospheres. 


2.20 REFRIGERATED SUITS 


Where individuals must move about, cold air may be blown 
into a suit or hood worn as a portable enclosure. The usual 
refrigeration methods may be used with insulated tubing to 
the suit. It may be difficult, however, to deliver air at a 
sufficiently low temperature. If compressed air is available, 
cold air may be delivered from a vortex tube worn on the suit. 
Suits of this type are commercially available. 


2.21 ENCLOSURES 


In certain hot industries, such as in steel mills, itis imprac- 
tical to control the heat from the process. If the operation is 
such that remote control is possible, an air conditioned booth 
or cab can be utilized to keep the operators reasonably com- 
fortable in an otherwise intolerable atmosphere. 


2.22 INSULATION 


If the source of heat is a surface giving rise to convection, 
insulation at the surface will reduce this form of heat transfer. 
Insulation by itself, however, will not usually be sufficient if 
the temperature is very high or if the heat content is high. 


REFERENCES 


2.1 U.S. Department of Health, Education and Welfare, 
PHS, CDC, NIOSH: The Industrial Environment—Its 
Evaluation and Control. Government Printing Office, 
Washington, DC (1973). 

2.2 U.S. Air Force: AFOSH Standard 161.2. 


2.3 National Board of Fire Underwriters: Pamphlet #86, 
Standards for Class A Ovens and Furnaces. 


2.4 Feiner, B.; Kingsley, L.: Ventilation of Industrial Ov- 


General Industrial Ventilation 2-17 


ens. Air Conditioning, Heating and Ventilating, pp. 82 
89 (December 1956). 


2.5 U.S. Department of Health and Human Services, PHS, 
CDC, NIOSH: Occupational Exposure to Hot Envi- 
ronments, Revised Criteria, 1986. 


2.6 American Conference of Governmental Industrial Hy- 
gienists, Inc.: 1997 Threshold Limit Values and Bio- 
logical Exposure Indices, p. 138, ACGIH, Cincinnati 
(1997). 


Chapter 3 
LOCAL EXHAUST HOODS 


3.1 INTRODUCTION ...............-..... 3-2 
3.2. CONTAMINANT CHARACTERISTICS ....... 3-2 
3.2.1 Inertial Effects ..............0.. 3-2 
3.2.2 Effect of Specific Gravity... 2... ...0.. 3-2 
3.2:3; “Wake Effects: s..:3-5:¢-9: 400 Sp eos ee Ae 3-2 
3.33. HOODTYPES...................2... 3-2 
3.3.1 Enclosing Hoods ................ 3-2 
3.3.2 ExteriorHoods ................. 3-2 
3.4 HOOD DESIGN FACTORS .............. 3-2 
3.4.1 Capture Velocity ........2.....2.. 3-6 
3.4.2 Hood Flow Rate Determination ........ 3-6 
3.4.3 Effects of Flanges and Baffles ......... 3-7 
3.4.4 Air Distribution ................. 3-7 
3.4.5 Rectangular and Round Hoods ......... 3-8 
3.4.6 Worker Position Effect... .......... 3-8 
3.5 HOODLOSSES.................... 3-15 
Figure 3-1 Hood Nomenclature Local Exhaust... .... . 3-3 
Figure 3-2 Effects of Specific Gravity ............ 3-4 
Figure 3-3 Enclosure and Operator/Equipment Interface .. 3-5 
Figure 3-4 Point Suction Source»... 2 ee ee 3-6 
Figure 3-5 Flow Rate as Distance From Hood ....... . 3-7 
Figure 3-6 Velocity Contours — Plain Circular Opening . . 3-8 
Figure 3-7 Velocity Contours — Flanged Circular Opening . 3-8 
Figure 3-8 Flow/Capture Velocity .............. 3-9 
Figure 3-9 Flow/Capture Velocity .............-. 3-10 
Figure 3-10 Flow/Capture Velocity .............. 3-1] 
Figure 3-11 Hood Ey pes! 50s 6 She ted cane oP ORE Shh 3-12 
Figure 3-12 Distribution Techniques — Slot Resistance and 


Hishebails 4.2% 2. Satie sired ats aad 3-13 


3.5.1 Simple Hoods ................. 3-16 
3.5.2 CompoundHoods ............... 3-16 
3.6 MINIMUMDUCT VELOCITY ............ 3-18 
3.7 SPECIAL HOOD REQUIREMENTS ......... 3-18 
3.7.1 Ventilation of Radioactive and High Toxicity 
Processes 2.425 2.25 20:4 shade wed 3-18 
3.7.2 Laboratory Operations. ............ 3-19 
3.8 PUSH-PULL VENTILATION .............. 3-19 
3.8.1. “BushJet: oa wb ke i ies wae BG 3-19 
3:8.2.. “PUILHOOd » x03 ain ect pickle panels LR eae 3-20 
3.8.3. Push-Pull System Design .. 2... 2.2... 3-21 
3.9 HOTPROCESSES ................0.. 3-21 
3.9.1 Circular High Canopy Hoods ..... 2... 3-21 
3.9.2 Rectangular High Canopy Hoods ...... . 3-22 
3.9.3 LowCanopy Hoods .............. 3-23 
REFERENCES bi eee genie 44s oe Gib HAD bh SLE os 3-23 
Figure 3-13 Distribution Techniques - Booth Canopy and 
Side-Drafts and Suspended Hoods ...... . 3-14 
Figure 3-14. | Worker Position Effect... .......02.. 3-15 
Figure 3-15 Airflow at the Vena Contracta ......... 3-16 
Figure 3-16 Hood Loss Factors ............-.. 3-17 
Figure 3-17. SimpleHood .................. 3-18 
Figure 3-18 CompoundHood ................ 3-18 
Figure 3-19 Jet Velocity Profile ............... 3-20 
Figure 3-20 Dimensions Used to Design High-Canopy Hoods 
for Hot Sources. 2... 2. ee 3-20 


3-2 Industrial Ventilation 


3.1 INTRODUCTION 


Local exhaust systems are designed to capture and remove 
process emissions prior to their escape into the workplace 
environment. The local exhaust hood is the point of entry into 
the exhaust system and is defined herein to include all suction 
openings regardless of their physical configuration. The pri- 
mary function of the hood is to create an air flow field which 
will effectively capture the contaminant and transport it into 
the hood. Figure 3-1 provides nomenclature associated with 
local exhaust hoods. 


3.2 CONTAMINANT CHARACTERISTICS 


3.2.1 Inertial Effects: Gases, vapors, and fumes will not 
exhibit significant inertial effects. Also, fine dust particles, 20 
microns or less in diameter (which includes respirable parti- 
cles), will not exhibit significant inertial effects. These mate- 
rials will move solely with respect to the air in which they are 
mixed. In such cases, the hood needs to generate an air flow 
pattern and capture velocity sufficient to control the motion 
of the contaminant-laden air plus extraneous air currents 
caused by room cross-drafts, vehicular traffic, etc. 


3.2.2 Effective Specific Gravity: Frequently, the location 
of exhaust hoods is mistakenly based on a supposition that the 
contaminant is "heavier than air" or "lighter than air." In most 
health hazard applications, this criterion is of little value (see 
Figure 3-2). Hazardous fine dust particles, fumes, vapors, and 
gases are truly airborne, following air currents, and are not 
subject to appreciable motion either upward or downward 
because of their own density. Normal air movement will 
assure an even mixture of these contaminants. Exception to 
these observations may occur with very hot or very cold 
operations or where a contaminant is generated at very high 
levels and control is achieved before the contaminant be- 
comes diluted. 


3.2.3 Wake Effects: As air flows around an object, a phe- 
nomenon known as "boundary layer separation" occurs. This 
results in the formation ofa turbulent wake on the downstream 
side of the object similar to what is observed as a ship moves 
through the water. The wake is a region of vigorous mixing 
and recirculation. If the object in question is a person who is 
working with, or close to, a contaminant-generating source, 
recirculation of the contaminant into the breathing zone is 
likely. An important consideration in the design of ventilation 
for contaminant control is minimizing this wake around the 
human body and, to the extent possible, keeping contaminant 
sources out of these recirculating regions (see also Section 
3.4.6.) 


3.3 HOOD TYPES 


Hoods may be of a wide range of physical configurations 
but can be grouped into two general categories: enclosing and 
exterior. The type of hood to be used will be dependent on the 


physical characteristics of the process equipment, the con- 
taminant generation mechanism, and the operator/equipment 
interface (see Figure 3-3). 


3.3.1 Enclosing Hoods: Enclosing hoods are those which 
completely or partially enclose the process or contaminant 
generation point. A complete enclosure would be a laboratory 
glove box or similar type of enclosure where only minimal 
openings exist. A partial enclosure would be a laboratory hood 
or paint spray booth. An inward flow of air through the 
enclosure opening will contain the contaminant within the 
enclosure and prevent its escape into the work environment. 


The enclosing hood is preferred wherever the process con- 
figuration and operation will permit. If complete enclosure is 
not feasible, partial enclosure should be used to the maximum 
extent possible (see Figure 3-3). 


3.3.2 Exterior Hoods: Exterior hoods are those which are 
located adjacent to an emission source without enclosing it. 
Examples of exterior hoods are slots along the edge of the tank 
or a rectangular opening on a welding table. 


Where the contaminant is a gas, vapor, or fine particulate 
and is not emitted with any significant velocity, the hood 
orientation is not critical. However, if the contaminant con- 
tains large particulates which are emitted with a significant 
velocity, the hood should be located in the path of the emis- 
sion. An example would be a grinding operation (see Chapter 
10, VS-80-11). 


If the process emits hot contaminated air, it will rise due to 
thermal buoyancy. Use of a side draft exterior hood (located 
horizontally from the hot process) may not provide satisfac- 
tory capture due to the inability of the hood-induced air flow 
to overcome the thermally induced air flow. This will be 
especially true for very high temperature processes such as a 
melting furnace. In such cases, a canopy hood located over 
the process may be indicated (see Section 3.9). 


A variation of the exterior hood is the push—pull system 
(Section 3.8). In this case, a jet of air is pushed across a 
contaminant source into the flow field of a hood. Contaminant 
control is primarily achieved by the jet. The function of the 
exhaust hood is to receive the jet and remove it. The advantage 
of the push—pull system is that the push jet can travel in a 
controlled manner over much greater distances than air can 
be drawn by an exhaust hood alone. The push—pull system is 
used successfully for some plating and open surface vessel 
operations but has potential application for many other proc- 
esses. However, the push portion of the system has potential 
for increasing operator exposure if not properly designed, 
installed, or operated. Care must be taken to ensure proper 
design, application, and operation. 


3.4 HOOD DESIGN FACTORS 


Capture and control of contaminants will be achieved by 


Local Exhaust Hoods 


SOURCE — 
Se 


CAPTURE UN 


VELOCITY  —=~ % 
SOURCE —- 


AIR VELOCITY AT ANY POINT IN FRONT OF 
OPENING NECESSARY TO OVERCOME OPPOSING AIR CURRENTS AND TC 
CAPTURE THE CONTAMINATED AIR AT THAT POINT BY CAUSING [7 


INTO THE HOOD. 


: FACE VELOCITY~ AIR: VELOCITY AT THE HOOD OPENING 


SLOT VELOCETY AIR VELOCITY 4 IGH THE OPENINGS IN A Si 
USED PRIMARILY AS A MEANS OF O3TAINING 


ACROSS THE FACKI OF THE HOOD 


Be PLENUM VELOCITY= AIR VELOCITY IN THE PLENUM. FOR GOOD AIR DISTRIBUTION 
WITH SLOT-TYPES OF HOODS, THE MAXIMUM PLENUM VELOCITY 
SHOULD BE 1/2 OF THE SLOT VELOCH 


DUCT VELOCITY~ AIR VELOCITY THROUGH THE DUCT CROSS SECTION. WHEN SOLID MATERIAL IS 
PRESENT IN THE AIR STREAM, THE DUCT V=ZLOCITY MUST BE EQUAL TO OR 
GREATER THAN THE MINIMUM AIR VELOCITY REQUIRED TO 


PARTICLES IN THE AiR STREA 


AMERICAN CONFERENCE | HOOD NOMENCLATURE 
OF GOVERNMENTAL, LOCAL EXHAUST 
HYGIENISTS ee 


3-3 


3-4 


Industrial Ventilation 


PAINT DIP > : > PAINT DIP 


LOCATION 


SOLVENT VAPORS IN HEALTH HAZARD CONCENTRATIONS ARE NOT APPRECIABLY HEAVIER THAN AIR. 
EXHAUST FROM THE FLOOR USUALLY GIVES FIRE PROTECTION ONLY. 


AMERICAN CONFERENCE | BRFFECTS OF 
OF GOVERNMENTAL SPECIFIC GRAVITY 


F INDUSTRIAL HYGIENISTS a 


Local Exhaust Hoods 3-5 


GB) enc LOSING HOOD | ; SLAB, 
BELT O JS 


OVO 


FOPPER < HOPPER / 
a enlenesecenem fineness Nonenenraanefianremnnnnfianernnrsnnalens 
GOOD WT “() QT RAD 
EEN Cl US E 4 


ENCLOSE THE OPERATION AS MUCH AS POSSIBLE. THE MORE COMPLETELY ENCLOSED THE 


SOURCE, THE LESS AIR REQUIRED FOR CONTROL. 


PROCESS | ( | PROCESS 


GOOD BAD 


DIREBG. TION Ol" AUK FLOW 
LOCATE THE HOOD SO THE CONTAMINANT IS REMOVED AWAY FROM THE BREATHING 
ZONE OF THE OPERATOR. 


AMERICAN CONFERENCE | ENCLOSURE AND OPERATOR/ 
f OF GOVERNMENTAL — | EQUIPMENT INTERFACE : 
| INDUS" TRIAL HYGIENISTS [ire——q-9¢__rours_ 


3-6 Industrial Ventilation 


TABLE 3-1. Range of Capture Velocities®1: 5) 
Condition of Dispersion of Contamination 


Example 


Capture Velocity, fpm 


Released with practically no velocity into quiet air. Evaporation from tanks; degreasing, etc. 50-100 

Released at low velocity into moderately still air. Spray booths; intermittent container filling; low speed 100-200 
conveyor transfers; welding; plating; pickling 

Active generation into zone of rapid air motion. Spray painting in shallow booths; barrel filling; 200-500 
conveyor loading; crushers 

Released at high initial velocity into zone at very rapid air motion. Grinding; abrasive blasting; tumbling 500-2000 

In each category above, a range of capture velocity is shown. The proper choice of values depends on several factors: 

Lower End of Range Upper End of Range 

1. Room air currents minimal or favorable to capture. 1. Disturbing room air currents. 

2. Contaminants of fow toxicity or of nuisance value only. 2. Contaminants of high toxicity. 

3. Intermittent, low production. 3. High production, heavy use. 

4. Large hood-large air mass in motion. 4. Small hood-local control only. 

the inward air flow created by the exhaust hood. Air flow ¢ The fact that the contaminant is under the influence of 


toward the hood opening must be sufficiently high to maintain 
control of the contaminant until it reaches the hood. External 
air motion may disturb the hood-induced air flow and require 
higher air flow rates to overcome the disturbing effects. 
Elimination of sources of external air motion is an important 
factor in achieving effective control without the need for 
excessive air flow and its associated cost. Important sources 
of air motion are 


¢ Thermal air currents, especially from hot processes or 
heat-generating operations. 


° Motion of machinery, as by a grinding wheel, belt 
conveyor, etc. 


¢ Material motion, as in dumping or container filling. 
¢ Movements of the operator. 


¢ Room air currents (which are usually taken at 50 fpm 
minimum and may be much higher). 


* Rapid air movement caused by spot cooling and heating 
equipment. 


The shape of the hood, its size, location, and rate of air flow 
are important design considerations. 


3.4.1 Capture Velocity: The minimum hood-induced air 
velocity necessary to capture and convey the contaminant into 
the hood is referred to as capture velocity. This velocity will 
be a result of the hood air flow rate and hood configuration. 


Exceptionally high air flow hoods (example, large foundry 
side-draft shakeout hoods) may require less air flow than 
would be indicated by the capture velocity values recom- 
mended for smal! hoods. This phenomenon may be ascribed 
to: 


° The presence of a large air mass moving into the hood. 


the hood for a much longer time than is the case with 
small hoods. 


° The fact that the large air flow rate affords considerable 
dilution as described above. 


Table 3-1 offers capture velocity data. Additional informa- 
tion is found in Chapter 10. 


3.4.2 Hood Flow Rate Determination: Within the bounds 
of flanges, baffles, adjacent walls, etc., air will move into an 
opening under suction from all directions. For an enclosure, 
the capture velocity at the enclosed opening(s) will be the 
exhaust flow rate divided by the opening area. The capture 
velocity at a given point in front of the exterior hood will be 
established by the hood air flow through the geometric surface 
which contains the point. 


As an example, for a theoretical unbounded point suction 
source, the point in question would be on the surface of a 
sphere whose center is the suction point (Figure 3-4). 


The surface area of a sphere is 4nX?. Using V = Q/A 


SURFACE OF 


SPHERE - 
POINT SUCTION 
7 SOURCE 
a i 
yo % 


i A \ » CAP TURE 
/ f CAP TURE. 
: f / VELOCITY 
Piece Eee 

iat 
| fi 

\ ee ye el 

\ / 

: Va 

Ne a 


FIGURE 3-4. Point suction source 


(Equation 1.3), the velocity at point X on the sphere’s surface 
can be given by 


Q = V(4nX?) = 12.57VX? [3.1] 


where: 
Q = air flow into suction point, cfm 
V = velocity at distance X, fpm 
A= 4nX2 = area of sphere, ft? 


X = radius of sphere, ft 


Similarly, if an unbounded line source were considered, the 
surface would be that of a cylinder and the flow rate (neglect- 
ing end effects) would be 


Q = V(2nXL2) = 6.28 VXL [3.2] 


where: 
L= length of line source, ft 


Equations 3.1 and 3.2 illustrate, on a theoretical basis, the 
relationship between distance, flow, and capture velocity and 
can be used for gross estimation purposes. In actual practice, 
however, suction sources are not points or lines, but rather 
have physical dimensions which cause the flow surface to 
deviate from the standard geometric shape. Velocity contours 
have been determined experimentally. Flow®) for round 
hoods, and rectangular hoods which are essentially square, 
can be approximated by 


Q = V(10X? +A) [3.3] 


where: 
Q = air flow, cfm 
V = centerline velocity at X distance from hood, fpm 


X= distance outward along axis in ft. (NOTE: equa- 
tion is accurate only for limited distance of X, 
where X is within 1.5 D) 


A= area of hood opening, ft? 


D = diameter of round hoods or side of essentially 
square hoods, ft 


Where distances of X are greater than 1.5 D, the flow rate 
increases less rapidly with distance than Equation 3.3 indi- 
cates,0439) 


It can be seen from Equation 3.3 that velocity decreases 
inversely with the square of the distance from the hood (see 
Figure 3-5.) 


Figures 3-6 and 3-7 show flow contours and streamlines 
for plane and flanged circular hood openings. Flow contours 
are lines of equal velocity in front of a hood. Similarly, 
streamlines are lines perpendicular to velocity contours. (The 
tangent to a streamline at any point indicates the direction of 
air flow at that point.) 


Flow capture velocity equations for various hood configu- 
rations are provided in Figures 3-8, 3-9, 3-10, and 3-11. 


Local Exhaust Hoods 3-7 


3.4.3 Effects of Flanges and Baffles: A flange is a sur- 
face at and parallel to the hood face which provides a barrier 
to unwanted air flow from behind the hood. A baffle is a 
surface which provides a barrier to unwanted air flow from 
the front or sides of the hood. 


If the suction source were located on a plane, the flow area 
would be reduced (1/2 in both cases), thereby decreasing the 
flow rate required to achieve the same velocity. A flange 
around a hood opening has the same effect of decreasing the 
required flow rate to achieve a given capture velocity. In 
practice, flanging can decrease flow rate (or increase velocity) 
by approximately 25% (see Figures 3-6, 3-7, and 3-11). For 
most applications, the flange width should be equal to the 
square root of the hood area ( VA). 


Baffles can provide a similar effect. The magnitude of the 
effect will depend on the baffle location and size. 


Figure 3-11 illustrates several hood types and gives the 
velocity/flow formulas which apply. 


A summary of other equations for hood velocity and the 
impact of cross-drafts on hood performance can be found in 
Reference 3.25. 


3.4.4 Air Distribution: Slot hoods are defined as hoods 
with an opening width-to-length ratio (W/L) of 0.2 or less. 
Slot hoods are most commonly used to provide uniform 


0.5 m/s NEEDED 
SOURCE ee 
C) et 
mw, 
be Xx -4 
exert 
2 m/s NEEDED 
} SN 
SOURCE | Sl \ 

( | 
eee 
| feo 
| Va 
bh 2X 


i . N ry NT 
LOCATION 
PLACE HOOD AS CLOSE TO THE SOURCE 
CONTAMINANT AS PO 
VOLUME VARIES WITH THE 
DISTANCE FROM THE SOURCE. 


THE 


FIGURE 3-5. Flow rate as distance from hood 


3-8 Industrial Ventilation 


i 
i H 
i “| 
i 
—+—L iS een eee ee 
AGS O} re) ‘ore 
2 rt) *y ~ ~ 
|p em 
( : 
a NA. 
fT 
A ea Se ee Lae 
i 
| 
i Jalen 
H 
ieee =e een 
50 100 


% OF DIAMETER 


FIGURE 3-6. Velocity contours — plain circular opening — % of opening 
velocity 


exhaust air flow and an adequate capture velocity over a finite 
length of contaminant generation, e.g., an open tank or over 
the face of a large hood such as a side-draft design. The 
function of the slot is solely to provide uniform air distribu- 
tion. Slot velocity does not contribute toward capture velocity. 
A high slot velocity simply generates high pressure losses. 
Note that the capture velocity equation (Figure 3-11) shows 
that capture velocity is related to the exhaust volume and the 
slot length, not to the slot velocity. 


Slot hoods usually consist of a narrow exhaust opening and 
a plenum chamber. Uniform exhaust air distribution across 
the slot is obtained by sizing slot width and plenum depth so 
that velocity through the slot is much higher than in the 
plenum. Splitter vanes may be used in the plenum; however, 
in most industrial exhaust systems, vanes are subject to cor- 
rosion and/or erosion and provide locations for material to 
accumulate. Adjustable slots can be provided but are subject 
to tampering and maladjustment. The most practical hood is 
the fixed slot and unobstructed plenum type. The design of 
the slot and plenum is such that the pressure loss through the 
slot is high compared with the pressure loss through the 
plenum. Thus, all portions of the slot are subjected to essen- 
tially equal suction and the slot velocity will be essentially 
uniform. 


There is no straightforward method for calculating the 
pressure drop from one end to the other of a slot-plenum 
combination. A very useful approximation, applicable to most 


hoods, is to design for a maximum plenum velocity equal to 
one-half of the slot velocity. For most slot hoods, a 2000 fpm 
slot velocity and 1000 fpm plenum velocity is a reasonable 
choice for uniformity of flow and moderate pressure drop. 
Centered exhaust take-off design results in the smallest prac- 
tical plenum size since the air approaches the duct from both 
directions. Where large, deep plenums are possible, as with 
foundry shake-out hoods, the slot velocity may be as low as 
1000 fpm with a 500 fpm plenum velocity. 


3.4.5 Rectangular and Round Hoods: Air distribution 
for rectangular and round hoods is achieved by air flow within 
the hood rather than by pressure drop as for the slot hood. The 
plenum (length of hood from face to tapered hood to duct 
connection) should be as long as possible. The hood take-off 
should incorporate a 60° to 90° total included tapered angle. 
Multiple take-offs may be required for long hoods. End 
take-off configurations require large plenum sizes because all 
of the air must pass in one direction. 


Figures 3-12 and 3-13 provide a number of distribution 
techniques. 


3.4.6 Worker Position Effect: The objective of industrial 
ventilation is to control the worker’s exposure to toxic air- 
borne pollutants in a safe, reliable manner. As one of the main 


% OF DIAMETER 


FIGURE 3-7. Velocity contours — flanged circular opening — % of opening 
velocity 


Local Exhaust Hoods 


| \ 
J 


(y we () 


SOURCE —~ 


im 
— SOURCE — 


FREELY SUSPENDED HOOD LARGE HOOD 


Q = VWUI0OX’ + A) LARGE HOCD, X SMALL~-~-MEASURE xX 
PERPENDICULAR TO HOOD FACE, NOT LESS 
THAN 2X FROM HOOD EDGE. 


rs 


SOURCE 
a 
! | 


HOOD ON BENCH OR FLOOR YD WITH WIDE FLANGE 


2 
Q = 5x + A) Q = V 0.75(10X" + A) 


SUSPENDED HOODS 
(SMALL SIDE-DRAFT HOODS) 


REQUIRED EXHAUST AIR FLOW, CFM. 

DISTANCE FROM HOOD FACE TO FARTHEST POINT OF CONTAMINANT RELEASE, FT. 

HOOD £ACE AREA, FT? . 

CAPTURE VELOCITY, FPM, AT DISTANCE x, 

NOTE: AIR FLOW RATE MUST INCREASE AS THE SQUARE OF DISTANCE OF 2CE FROM THE HOOD 
BAFFLING BY FLANGING OR BY PLACING ON BENCH, FLOOR, ECT. HAS A BENEFICIAL EFF 


Q 
x 
A 
V 


toa ou 


5° MINIMUM 


CANOPY HOOD 
Q = 1.4 PDV(P=PERIMETER OF TANK, FT). 
OT RECOMMENDED IF WORKERS WUST BEND OVER SOURCE. Vv RANGES 
FROM 50 TO 500 FPM DEPENDING ON CROSSDRAFTS. SIDE CURTAINS ON TWO OR THREE SIDES TO 
CREATE A SEMI-BOOTH OR BOOTH ARE DESIRABLE. 


\MERICAN CONFERENCE 
OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


FLOW/CAPTURE VELOCITY 


= a = a 


3-9 


3-10 Industrial Ventilation 


FREELY SUSPENDED SLOT 
Q = 3.7LVX 


- FLANGE (WIDTH = vA”) Q = REQUIRED EXHAUST FLOW RATE, CFM 
= DISTANCE, HOOD se FO) FAR TI HES T 
POINT OF SOURCE (USUALLY ON 
CENTERLINE OF oon FT, 
CAPTURE VELOCITY AT DISTANCE X, FPM, 
= LENGTH, OF HOOD, SLOT, TABLE, TANK, ETC, FT. 
= WIDTH, CF TABLE, TANK, ETC., F lie 
= HOOD FACE AREA, FT. 


FLANGED SLOT 
Q = CLW 


SLOT ON TANK FLANGED SLOT 
Q = CLW Q = CLW 
S~HALF Q IN EACH SLOT IF 
SLOTS ON BOTH SIDES 


560, 


AMERI : AN CONFERENCE | | eee cates: 
ae OW/CAPTURE VELOCIT’ 
sn cee ata FLOW/CAPTURE VELOCITY 


B INDUS a HYGIENISTS Kas —gg ee / 


Local Exhaust Hoods 3-11 


ming 


SIME AS TO BOOTH SIMILAR TO SUSPENDED HOOD 


DOWNDRAFT HOODS 


NO? RECOMMENDED FOR HOT OR HEAT~PR ODUCING OPERATIONS IF DOWNDRAFT AREA 
S LARGE, SEE "CAPTURE VELOCITY TION 


ANGLE z 


IF DESIRED 


rrr ec 


BOOTHS TPE. <Q ODS 


AREA, FT?;  V=FACE VELOCITY, FPM). 
ES OPTIONAL FOR AIR RIBUTION; NOT REQUIRED IF A WATER WALL BOOTH 
OTHER MEANS FOR DIST IN IS PROVIDES 
S FROM 4 INCHES TO 8 INCHES, DEPENDING ON SIZE OF BOOTH. 
ES FROM 6 INCHES TO 12 INCHES, DEPENDING ON SIZE OF BOOTH. 
EF THE NUMBER OF PANELS WITH SIZE OF BOOTH. 


ONE ENG 


2OVERNMENTAL 
US TRIAL a edboes 


PLOW/CAPTURE VELOCITY 


3-12 Industrial Ventilation 


] 
HOOD TYPE DESCRIP TION ASPECT RATIO,W/L AIR FLOW 


0.2 OR LESS 


FLANGED SLOT 0.2 OR LESS 


0.2 OR GREATER 
AND ROUND 


Q = V(10X’ +A) 


PLAIN OPENING 


0.2 OR GREATER Q 


2 
ete ane = 0.75V(10X +A) 


FLANGED OPENING 


TO SUIT WORK 


1.4 PVD 
SEE FIG. VS-—99-03 
TO SUIT WORK P = PERIMETER 
D = HEIGHT 
ABOVE WORK 


CANOPY 


PLAIN MULTIPLE 
SLOT OPENING 
2 OR MORE SLOTS 


2 
0.2 OR GREATER Q = V(10X" +A) 


FLANGED MULTIPLE 
SLOT OPENING 
2 OR MORE SLOTS 


2 
0.2 OR GREATER Q = 0.75V(10X +A) 


AMERICAN CONFERENCE 
, OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


HOOD TYPES 


Local Exhaust Hoods 3-13 


INSIDE RADIUS MORE IMPORTAN] 


THAN OUTSIDE 


SLOT VELOCITY 2000 FPM MAX. PLENUM 


OR HIGHER. - VELOCITY=1/2 SLOT VELOCITY 


tp | 


| 
(| 
| 


ij 


SLOPE FOR DRAINING !S DESIRABLE : 
SLOPE DOES NOT AID IN DISTRIBUTION 


DISTRIBUTION BY SLOT RESISTANCE 


DISTRIBUTION BY FISH TAIL 


WITH LOW PLENUM VELOCITIES AND HIGH SLOT VELOCITIES, GOCD DISTRIBUTION IS OBTAINED. 
SLOTS OVER 10 FEET TO 12 FEET IN LENGTH USUALLY NEED MULTIPLE TAKE-OFFS. 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
INDUSTRIAL tlYGIEN:STS 


DISTRIBUTION TECHNIQUES 


3-14 Industrial Ventilation 


ON BY BAFFLES 


IBUTION B 


BOOTH CANO 


{ SAME PRINCIPLES APPLY TO CANOPY TYPE ) 


30° to 45° 


DISTRIBUTION BY DISTRIBUTION BY TAPER 
SLOT (OR BAFFLES) 


AMERICAN CONFERENCE 
a GONE ne NTAL 
INDUSTRIAL is | : NE [STs 


DISTRIBUTION TECHNIQUES 


engineering controls, local exhaust ventilation is designed to 
be near the point of contaminant generation. Often, considera- 
tion is not given to how the workers will position themselves 
with respect to the air flow. Studies°*?> show that the posi- 
tion of the worker with respect to the flow direction is an 
important parameter in determining the breathing zone con- 
centration. 


Figure 3-14, Position 2, shows a worker oriented with his 
back to the air flow. Immediately downstream of the worker, 
a zone of reverse flow and turbulent mixing occurs due to 
boundary layer separation. Contaminant released into this 
region (e.g., from a hand-held or proximal source) will be 
mixed into the breathing zone resulting in exposure. Figure 
3-14, Position 1, shows a worker oriented at 90° to the flow 
direction; here, the reverse flow zone forms to the side and 
there is less opportunity for the entrainment of contaminant 
into the breathing zone. 


Studies suggest that this phenomenon is important when 
large booth-type hoods are employed or in situations where 
there is a reasonably uniform air flow. Exposure studies®:!”) 
using a tailor’s mannequin to simulate an operator in a booth- 
type hood used for the transfer of powders showed, in all 
cases, that exposures for Position 1 were less than those in 
Position 2 by at least a factor of 2000. 


A second case study reported women who used a spray 
and brush application of a chloroform-based adhesive were 
significantly exposed despite working in a ventilated booth. 
A 50% reduction in exposure was found when the workers 
stood side-on to the air flow (Position 1). Subsequent modifi- 
cation of spray practices resulted in a determination that a 30° 
angle to the air flow and holding the nozzle in the downstream 
hand seemed optimal. No alterations to the actual design or 
air flow of the booth were needed to achieve acceptable 
exposure levels. 


The preceding discussion assumes that the worker is not in 
the wake of an upstream object and that the contaminant 
source has negligible momentum. In cases where the contami- 
nant source has significant momentum (e.g., high-pressure 
compressed air paint spray operations), the effect of position 
on exposure may be reversed — i.e., Position | in Figure 3-14 
may produce higher exposures. This is associated with the 
deflection of the spray upstream of the worker and subsequent 
recirculation through the breathing zone. Further research and 
field studies are needed to evaluate the tendency for reverse 
flow to occur in more complex situations. Although the im- 
portance of boundary layer separation effects with smaller 
local exhaust hoods has not been thoroughly explored, three 
studies &!'3.15) suggest that the 90° orientation is beneficial 
even in this instance. It is recommended that the side orienta- 
tion (i.e., Position 1) be the preferred orientation in situations 
where feasible. Down-draft configurations may provide simi- 
lar benefits under certain conditions. 


It is recommended that the side orientation (i.e., Position 


Local Exhaust Hoods 3-15 


oe — 
-- ee : ® Source - Sea cna ae 
ie we 
Airflow —te= 
cass gat Dace Seren . 
sae fF 
a 
a Zs oy “ 
Be — ~ 
am 2 Ny N 
- f | SNe oy 
neers an ee 
Airflow —hke eee, ® Source 
y 4 ( ; wo 
Pe oe # ¢ 
eee Sa 33 we 
Position #2 


FIGURE 3-14. Worker position effect 


1) be investigated as a preferred work practice where feasible. 
It is important to assess the exposure with personal sampling 
pumps to confirm the benefits of one position versus another 
as other factors may complicate the issue. 


3.5 HOOD LOSSES 


Plain duct openings, flanged duct openings, canopies, and 
similar hoods have only one significant energy loss. As air 
enters the duct, a vena contracta is formed and a small energy 
loss occurs first in the conversion of static pressure to velocity 
pressure (see Figure 3-15.) As the air passes through the vena 
contracta, the flow area enlarges to fill the duct and velocity 
pressure converts to static pressure. At this point, the uncon- 
trolled slow down of the air from the vena contracta to the 
downstream duct velocity results in the major portion of the 
entry loss. The more pronounced the vena contracta, the 
greater will be the energy loss and hood static pressure. 


Compound hoods are hoods which have two or more points 
of significant energy loss and must be considered separately 
and added together to arrive at the total loss for the hood. 
Common examples of hoods having double entry losses are 
slot-type hoods and multiple-opening, lateral draft hoods 
commonly used on plating, paint dipping and degreasing 
tanks, and foundry side-draft shakeout ventilation. 


The hood entry loss (h,) can be expressed, therefore, in 
terms of hood loss coefficients (F;) which, when multiplied 
by the slot or duct velocity pressure (VP), will give the entry 
loss (h,) in inches of water. The hood static pressure is equal 


3-16 Industrial Ventilation 


\ ( 


FIGURE 3-15. Air flow at the vena contracta 


to the hood entry loss plus the velocity pressure in the duct. 
The hood entry loss represents the energy necessary to over- 
come the losses due to air moving through and into the duct. 
The velocity pressure represents the energy necessary to 
accelerate the air from rest to duct velocity (see Chapter 1, 
Section 1.6, "Acceleration of Air and Hood Entry Losses.") 
This may be expressed as: 


SP, =h, + VP, [3.4] 
SP, = (F,)(WP,) + (Fy)(WP3) + VPy [3.5] 
where: 
h, = overall hood entry loss = h, + hg, "wg 
h, = slot or opening loss = (F,)(VP;), "wg 
hg= duct entry loss = (Fy)(VP,), "wg 
F, = loss coefficient for slot 
Fy= loss coefficient for duct entry 
VP, = slot or opening velocity pressure, "wg 
VP, = duct velocity pressure, "wg 


One exception can occur when the slot velocity (or other hood 
entry velocity is higher than is the duct velocity. In such case, 
the acceleration velocity pressure used in determining SP is 
the higher slot or opening velocity pressure. 


Figures 3-16 and 5-13 give hood entry loss coefficients for 
several typical hood types. 


3.5.1 Simple Hoods: A simple hood is shown in Figure 
3-17. If the hood face velocity for such a simple hood is less 
than 1000 fpm, h, will be negligible and the loss will be 
dependent on h, only. If the hood face velocity is greater than 
1000 fpm, both h, and h, should be considered. Face velocities 
greater than 1000 fpm will usually only occur with relatively 
small hood face areas (0.25 to 0.50 ft). 


EXAMPLE PROBLEM 


Given: Face Velocity (V,) = 8 = 250 fpm 
f 


Duct Velocity (V,) -2 = 3000 fpm 
f 


Vv 2 
VPq = a = 0.56 “wg 
4005 
Fq = 0.25 as shown in Figure 5-12 
SPh = ha + VPa 
= (0.25)(0.56) + 0.56 
= 0.70 “wg 


3.5.2 Compound Hoods: Figure 3-18 illustrates a double 


entry loss hood. This is a single slot hood with a plenum and 
a transition from the plenum to the duct. The purpose of the 


Local Exhaust Hoods 3-17 


HOCD TYPE | DESCRIPTION HOOD ENTRY LOSS (F , 


FLANGED OPENING 


TAPER OR CONE 
HOOD 


SEE CHAPTER 


BELL MOUTH 
INLET 


ORIFICE SEE CHAPTER 10 


(STRAIGHT TAKEOFF 


TYPICAL GRINDING 


rig (TAPERED TAKEOFF) 


0.40 


AMERICAN CONFERENCE 
: OF GOVERNMENTAL 
INDUSTRIAL. HYGIENISTS 


HOOD LOSS FACTORS 


3-18 Industrial Ventilation 


/ 


FIGURE 3-17. Simple hood 


plenum is to give uniform velocity across the slot opening. 
Air enters the slot, in this case a sharp-edged orifice, and loses 
energy due to the vena contracta at this point. The air then 
continues through the plenum where the greater portion of the 
slot velocity is retained because the air stream projects itself 
across the plenum ina manner similar to the "blowing" supply 
stream shown in Figure |-7.(The retention of velocity in the 
plenum is characteristic of most local exhaust hoods because 
of the short plenum length.) In the case of very large hoods or 
exhausted closed rooms, however, the velocity loss must be 
taken into account. Finally, the air converges into the duct 
through the transition where the second significant energy 
loss occurs. For this type hood, both h, and h, must be 
considered. 


EXAMPLE PROBLEM 
Given: Slot Velocity (V,) = 2000 fpm 
Duct Velocity (V4) = 3500 fpm 


(Va is greater than Vs; therefore, use VPq as 
the acceleration VP) 


Vv 2 
a = 0.25 "w 
Me (asta) 7 


F, for slot = 1.78 from Figure 5-12 


Fy = 0.25 as shown in Figure 5-13 
SP, = hs + hg + VPa 
= (1.78)(0.25)+ (0.25) (0.76) + 0.76 
= 1.40 "wg 


3.6 MINIMUM DUCT VELOCITY 


The velocity pressure, VPy, utilized to determine hood 
losses in the previous examples is determined from the air 
velocity in the duct immediately downstream of the hood to 
duct connection. This velocity is determined by the type of 


material being transported in the duct. 


For systems handling particulate, aminimum design veloc- 
ity is required to prevent settling and plugging of the duct. On 
the other hand, excessively high velocities are wasteful of 
power and may cause rapid abrasion of ducts.6!432)) Mini- 
mum recommended design velocities are higher than theoreti- 
cal and experimental values to protect against practical 
contingencies such as: 


1. Plugging or closing one or more branch will reduce the 
total flow rate in the system and correspondingly will 
reduce the velocities in at least some sections of the 
duct system. 


2. Damage to ducts, by denting for example, will increase 
the resistance and decrease the flow rate and velocity 
in the damaged portion of the system. 


. Leakage of ducts will increase flow rate and velocity 
downstream of the leak but will decrease air flow 
upstream and in other parts of the system. 


Ww 


4. Corrosion or erosion of the fan wheel or slipping of a 
fan drive belt will reduce flow rates and velocities. 


5. Velocities must be adequate to pick up or re-entrain 
dust which may have settled due to improper operation 
of the exhaust system. 


The designer is cautioned that for some conditions such as 
sticky materials, condensing conditions in the presence of 
dust, strong electrostatic effects, etc., velocity alone may not 
be sufficient to prevent plugging, and other special measures 
may be necessary. 


Some typical duct velocities are provided in Table 3-2. The 
use of minimum duct velocity is treated in detail in Chapter 5S. 


3.7 SPECIAL HOOD REQUIREMENTS 


3.7.1 Ventilation of Radioactive and High Toxicity 
Processes: Ventilation of radioactive and high toxicity 
processes requires a knowledge of the hazards, the use of 
extraordinarily effective control methods, and adequate main- 
tenance which includes monitoring. Only the basic principles 


Rr rye ry 


FIGURE 3-18. Compound hood 


TABLE 3-2. Range of Minimum Duct Design Velocities 


Nature of Contaminant Examples 


Local Exhaust Hoods 3-19 


Design Velocity 


Vapors, gases, smoke 


Fumes 
Very fine light dust 
Dry dusts & powders 


Average industrial dust 


Heavy dusts 


Heavy or moist 


All vapors, gases, and smoke 


Welding 
Cotton lint, wood flour, litho powder 


Fine rubber dust, Bakelite molding powder dust, jute lint, cotton dust, 
shavings (light), soap dust, leather shavings 


Grinding dust, buffing lint (dry), wool jute dust (shaker waste), coffee 
beans, shoe dust, granite dust, silica flour, general material handling, 
brick cutting, clay dust, foundry (general), limestone dust, packaging and 
weighing asbestos dust in textile industries 


Sawdust (heavy and wet), metal turnings, foundry tumbling barrels and 
shake-out, sand blast dust, wood blocks, hog waste, brass turnings, cast 
iron boring dust, lead dust 


Lead dusts with small chips, moist cement dust, asbestos chunks from 
transite pipe cutting machines, buffing lint (sticky), quick-lime dust 


Any desired velocity 
(economic optimum velocity 
usually 1000-2000 fpm) 


2000-2500 
2500-3000 
3000-4000 


3500-4000 


4000-4500 


4500 and up 


can be covered here. For radioactive processes, reference 
should be made to the standards and regulations of the nuclear 
regulatory agencies. 


Local exhaust hoods should be of the enclosing type with 
the maximum enclosure possible. Where complete or nearly 
complete enclosure is not possible, control velocities from 50 
to 100% higher than the minimum standards in this manual 
should be used. If the enclosure is not complete and an 
operator must be located at an opening, such as in front of a 
laboratory hood, the maximum control velocity should not 
exceed 125 fpm. Air velocities higher than this value will 
create eddies in front of the operator which may pull contami- 
nant from the hood into the operator’s breathing zone. Re- 
placement air should be introduced at low velocity and in a 
direction that does not cause disruptive cross drafts at the hood 
opening. 


3.7.2 Laboratory Operations: Glove boxes should be 
used for high-activity alpha or beta emitters as well as highly 
toxic and biological materials. The air locks used with the 
glove box should be exhausted if they open directly to the 
room. 


For low-activity radioactive laboratory work, a laboratory 
fume hood may be acceptable. For such hoods, an average 
face velocity of 80-100 fpm is recommended. See Section 
10.35, VS-35-01, -02, -04, and -20. 


For new buildings, it is frequently necessary to estimate the 
air conditioning early — before the detailed design and equip- 
ment specifications are available. For early estimating, the 
guidelines provided in Section 10.35 for hood air flow and 
replacement air flow can be used. 


3.8 PUSH—PULL VENTILATION 


Push-pull ventilation consists of a push nozzle and an 
exhaust hood to receive and remove the push jet. Push-pull 
is used most commonly on open surface vessels such as 
plating tanks@?” but may be effectively used elsewhere (see 
VS-70-10). The advantage of push~pull is that a push jet will 
maintain velocity over large distances, 20-30 ft or more, 
whereas the velocity in front of an exhaust hood decays very 
rapidly as the distance from the hood increases. Properly used, 
the push jet intercepts contaminated air and carries it relatively 
long distances into the exhaust hood, thus providing control 
where it may be otherwise difficult or impossible. 


3.8.1 Push Jet: Ambient air is entrained in the push jet and 
results in a jet flow at the exhaust hood several times greater 
than the push nozzle flow rate. The jet velocity will decay with 
distance from the nozzle. The entrainment ratio for a long thin 
slot-(or pipe) type nozzle may be approximated by:°) 


Qs L412 [=| +044 [3.6] 
Q, 


ie} 


The velocity ratio may be approximated by:6?9) 


Vy 42 
Vo i [3.7] 
° [> +041 
by 
where: 


Q, = the push nozzle supply flow 
Q 
V, 
V, = the peak push jet velocity at a distance x 


the jet flow rate at a distance x from the nozzle 


“ 


the push nozzle exit air velocity 


°° 


3-20 Industrial Ventilation 


+y PEAK VELOCITY 
6) 4 

_ JET 
y CENTERLINE 


VELOCITY —> 


(c) FREE PLANE JET 


PEAK 
VELOCITY 


ify WALL 
OR 
SURFACE 


VELOCITY — 


(b) PLANE WALL JET 


FIGURE 3-19. Jet velocity profile 


a = a coefficient characteristic of the nozzle (0.13 
for slots and pipes) 


x = distance from the nozzle 
b, = the slot width* 


[*If the nozzle is freely suspended (free plane jet), b, 
is equal to one-half the total slot width. If the nozzle 
is positioned on or very near a plane surface (wall jet), 
b, is equal to the full slot width. For pipes with holes, 
b, is the width of a slot with equivalent area.] 


Typical jet velocity profiles are shown in Figure 3-19. 


Obstructions in the jet path should be minimized near the 
jet. Objects with small cross-sections, such as parts hangers, 
will cause serious problems; however, large flat surface ob- 
jects should be avoided. At further distances from the nozzle 
where the jet has expanded, larger objects may be acceptable 
if they are located within the jet. 


The nozzle may be constructed as a long thin slot, a pipe 
with holes or individual nozzles. The total nozzle exit area 
should not exceed 50% of the nozzle plenum cross-sectional 


area to assure even flow distribution. Slot width can range 
from 0.125—0.25 inch for short push length such as plating 
tanks (4-8 ft). Hole size should be 0.25 inch on 3 to 8 diameter 
spacing. The nozzle momentum factor, which is proportional 
to nozzle exit flow per foot of nozzle length times nozzle exit 
velocity (Q, x V,), must be sufficient to result in an effective 


jet but not so strong that the exhaust hood is overpowered. A 


QoV_ range should be approximately 50,000—-75,000 per foot 
of nozzle length for short distances of 4-8 feet. 


3.8.2 Pull Hood: The pull hood will accept and remove the 
push jet flow. The same design considerations regarding flow 
distribution, hood entry losses, etc., used for a normal pull- 
only hood should be used. The hood pull flow should be 
approximately |.5—2.0 times the push flow which reaches the 
hood. If design criteria specifying pull flow rate are not 
available, Equation 3.6 can be used. 


The hood opening height should be the same as the width 
of the expanded jet, if possible. However, smaller opening 
heights are acceptable if the hood flow rate meets the |.5—2.0 
times jet flow criteria. 


HYPOTHETICAL 


POINT SOURCE \ 


FIGURE 3-20. Dimensions used to design high-canopy hoods for hot 
sources (Ref. 3.24) 


Each push-pull application will necessitate special atten- 
tion. Wherever possible, a pilot system should be evaluated 
prior to final installation. 


3.8.3 Push—Pull System Design: Specific design criteria 
have been developed experimentally for plating, cleaning, or 
other open surface vessels and are provided in VS-70-10,VS- 
70-11, and VS-70-12. Where such specific design criteria are 
not available, the criteria provided in Sections 3.8.1 and 3.8.2 
can be used. When designing with Equation 3.7, a push jet 
velocity (V,) of 150-200 fpm at the exhaust hood face should 
be specified. 


3.9 HOT PROCESSES 


Design of hooding for hot processes requires different 
considerations than design for cold processes.°*4) When sig- 
nificant quantities of heat are transferred to the air above and 
around the process by conduction and convection, a thermal 
draft is created which causes an upward air current with air 
velocities as high as 400 fpm. The design of the hood and 
exhaust rate must take this thermal draft into consideration. 


3.9.1 Circular High Canopy Hoods: As the heated air 
rises, it mixes turbulently with the surrounding air. This 
results in an increasing air column diameter and volumetric 
flow rate. The diameter of the column (see Figure 3-20) can 
be approximated by: 


D-=05Kr" [3.8] 


where: 

D,= column diameter at hood face 

X,= y+z=the distance from the hypothetical point 

source to the hood face, ft 

y = distance from the process surface to the hood 
face, ft 
distance from the process surface to the hypo- 
thetical point source, ft 


N 
ul 


"z" can be calculated from: 


z= (2 Ag yrs 13.9] 


where: 
A, = diameter of hot source, ft. 


The velocity of the rising hot air column can be calculated 
from: 


(At)?*? 
V, = 8(A,)°S om [3.10] 
c 
where: 
V,= velocity of hot air column at the hood face, fpm 
A, = area of the hot source, ft? 
At= the temperature difference between the hot 


Local Exhaust Hoods 3-21 


source and the ambient air, F 
X,= y +z=the distance from the hypothetical point 
source to the hood face, ft. 


The diameter of the hood face must be larger than the 
diameter of the rising hot air column to assure complete 
capture. The hood diameter is calculated from: 

D, =D, = 0.8y [3.11] 
where: 

D, = diameter of the hood face, ft 


Total hood air flow rate is 
Qt =VyAc + V, (Ag Ac) [3.12] 


where: 
Q, = total volume entering hood, cfm 
V; = velocity of hot air column at the hood face, fpm 
A, = area of the hot air column at the hood face, ft? 


V, = the required air velocity through the remaining 
hood area, fpm 


A; = total area of hood face, ft? 
EXAMPLE PROBLEM 

Given: 4.0 ft diameter melting pot (Da) 

1000 F metal temperature 

100 F ambient temperature 

Circular canopy hood located 10 ft above pot (y) 
Calculate x,: 
Xe = ytz=y + (2Ds)"1% 
Xe = 10 + (2 x 4)1-198 
Xc = 10.7 ft 


Calculate the diameter of the hot air column at the hood 
face: 


De=05x 
De = 0.5(20.7)° 8 
De = 7.2 ft 


Calculate the velocity of the hot air column at the hood face: 


(aty?-*2 


V, = 8(A,)°** oy 


As = 0.25nD¢* 
As = 0.25n(4.2)" 
As = 12.6 ft? 


At = 1000 — 100 = 900 F 


3-22 Industrial Ventilation 


0.33 (900)? 


V; = 8(126) 207" 
7 (17.4) 

Vv; = (8)(231) O73) 

V, = 151 fpm 


Calculate diameter of hood face: 


Dr = De + 0.8y 
Dr= 7.2 = 0.8" 
Dr= 15.2 ft 


Calculate total hood airflow rate 
Qr = ViAc + Vr(Ar Ac) 
Ac = 0.25nDc* 
Ac = 0.25n(7.2)* 
Ac= 41 ft? 
Ay = 0.25nD¢" 
Ay = 0.25n(7.2)" 
Ar = 181 # 
Qr = 151(41) + 100(181 — 41) 
Qr= 10,290 cfm 


3.9.2 Rectangular High Canopy Hoods: Hot air col- 
umns from sources which are not circular may be better 
controlled by a rectangular canopy hood. Hood air flow 
calculations are performed in the same manner as for circular 
hoods except the dimensions of the hot air column at the hood 
(and the hood dimensions) are determined by considering 
both the length and width of the source. Equations 3.8, 3.9, 
and 3.11 are used individually to determine length and width 
of the hot air column and the hood. The remaining values are 
calculated in the same manner as for the circular hood. 


EXAMPLE PROBLEM 
Given: 2.5 ft x 4 ft rectangular melting furnace 
700 F metal temperature 
80 F ambient temperature 


Rectangular canopy hood located 8 ft above 
furnace (y) 


Calculate X, for each furnace dimension. 
X25 =y + Z25=y + (2Ds2.5)'"° 
=8+(2x2.5)' 1% 
= 14.2 ft 


Xea = B+ (2x 41192 
= 18.7 ft 
Calculate the width of the hot air column at the hood face. 
Deas = 0.5 Xc2.5°°8 
= 0.5(14.2)° 88 
= 5.2 ft 
Dea.o = 0.5(18.7)° 
= 6.6 ft 
Calculate the velocity of the hot air column at the hood face. 


0.33 (Aty?*? 
(xX, 


Vi = 8 (Ag) 
As=2.5x4= 10 ft 

At = 700 — 80 = 620 F 

Xe = Xc2.5 = 14.2 ft 

Note: X95 is used rather than X.49 as it is smaller and 


as such will yield a slightly larger V, which results in a 
margin of safety. 


v =8(10)022 (620) — 
10) (14.2)°25 


=8 (21) moe 


= 132 fpm 
Calculate hood face dimensions. 
Hood width = Dc2.5 + 0.8y 
= (5.2) + 0.8(8) 
= 11.6 ft 
Hood length = Dc4.0 + 0.8y 
= 6.6 + 0.8(8) 
= 13.0 ft 
Calculate the total hood air flow rate. 
Qi = ViAc + Vi (At — Ac) 
Ac = (Dc2.5)(De4.0) 
= (5.2)(6.6) 
= 34 ft? 
A¢ = (hood length) (hood width) 
= (11.6)(13.0) 
= 151 ft” 


Qt = (151)(34) + 100(151 - 34) 
= 5134 + 11,700 
= 16,834 cfm 


3.9.3 Low Canopy Hoods: If the distance between the 
hood and the hot source does not exceed approximately the 
diameter of the source or 3 ft, whichever is smaller, the hood 
may be considered a low canopy hood. Under such conditions, 
the diameter or cross-section of the hot air column will be 
approximately the same as the source. The diameter or side 
dimensions of the hood therefore need only be 1 ft larger than 
the source. 


The total flow rate for a circular low canopy hood is 


Q, = 4.7 (D,)? (at)? [3.13] 


where: 
Q, = total hood air flow, cfm 
D,; = diameter of hood, ft 
At= difference between temperature of the hot 
source, and the ambient, F. 


The total flow rate for a rectangular low hood is 


Qa =62 bts ato-42 
L 
where: 
Q, = total hood air flow, cfm 
L = length of the rectangular hood, ft 
b= width of the rectangular hood, ft 
At= difference between temperature of the hot 
source and the ambient, F. 
REFERENCES 


3.1. Brandt, A.D.: Industrial Health Engineering. John 
Wiley and Sons, New York (1947). 


3.2. Kane, J.M.: Design of Exhaust Systems. Health and 
Ventilating 42:68 (November 1946). 


3.3. Dalla Valle, J.M.: Exhaust Hoods. Industrial Press, 
New York (1946). 


3.4. Silverman, L.: Velocity Characteristics of Narrow Ex- 
haust Slots. J. Ind. Hyg. Toxicol. 24:267 (November 
1942). 


3.5. Silverman, L.: Center-line Characteristics of Round 
Openings Under Suction. J. Ind. Hyg. Toxicol. 24:259 
(November 1942). 


3.6. Piney, M.; Gill, F.; Gray, C.; et al.: Air Contaminant 
Control: the Case History Approach — Learning From the 
Past and Looking to the Future. In: Ventilation '88, J. H. 
Vincent, Ed., Pergammon Press, Oxford, U.K. (1989). 


3.7. Ljungqvist, B.: Some Observations on the Interaction 


3.8. 


3: 


3:12. 


3.14. 


3.20. 


3.21. 


3.22. 


3.23. 


Local Exhaust Hoods 3-23 


Between Air Movements and the Dispersion of Pollu- 
tion. Document D8:1979. Swedish Council for Build- 
ing Research, Stockholm, Sweden (1979). 


Kim, T.; Flynn, M.R.: Airflow Pattern Around a 
Worker in a Uniform Freestream. Am. Ind. Hyg. As- 
soc. J. 52:(7):187-296 (1991). 


George, D.K.; Flynn, M.R.; Goodman, R.: The Impact 
of Boundary Layer Separation on Local Exhaust De- 
sign and Worker Exposure. Appl. Occup. Env. Hyg. 
5:501-509 (1990). 


. Heriot, N.R.; Wilkinson, J.: Laminar Flow Booths for 


the Control of Dust. Filtration and Separation 
16:2:159-164 (1979). 


. Flynn, M.R.; Shelton, W.K.: Factors Affecting the 


Design of Local Exhaust Ventilation for the Control 
of Contaminants from Hand-held Sources. Appl. Oc- 
cup. Env. Hyg. 5:707-714 (1990). 


Tum Suden, K.D.; Flynn, M.R.; Goodman, R.: Com- 
puter Simulation in the Design of Local Exhaust 
Hoods for Shielded Metal Arc Welding. Am. Ind. Hyg. 
Assoc. J., 51(3): 115-126 (1990). 


. American Welding Society: Fumes and Gases in the 


Welding Environment. F. Y. Speight and H. C. Camp- 
bell, Eds. AWS, Miami, FL (1979). 


American Society of Mechanical Engineers: Power 
Test Code 19.2.4: Liquid Column Gages. ASME 
(1942). 


. Hemeon, W.C.L.: Plant and Process Ventilation. In- 


dustrial Press, New York (1963). 


. Alden, J.L.: Design of Industrial Exhaust System. 


Industrial Press, New York (1939). 


. Rajhans, G.S.; Thompkins, R.W.: Critical Velocities 


of Mineral Dusts. Canadian Mining J. (October 1967). 


. Djamgowz, O.T.; Ghoneim, S.A.A.: Determining the 


Pick-Up Air Velocity of Mineral Dusts. Canadian 
Mining J. (July 1974). 


. Baliff, J.L.; Greenburg, L.; Stern, A.C.: Transport 


Velocities for Industrial Dusts — An Experimental 
Study. Ind. Hyg. Q. (December 1948). 


Dalla Valle, J.M.: Determining Minimum Air Veloci- 
ties for Exhaust Systems. Heating, Piping and Air 
Conditioning (1932). 


Hatch, T.F.: Economy in the Design of Exhaust Sys- 
tems. 


Hughes, R.T.: Design Criteria for Plating Tank Push- 
Pull Ventilation. In: Ventilation '85. Elsevier Press, 
Amsterdam (1986). 


Baturin, V.V.: Fundamentals of Industrial Ventilation. 
Pergamon Press, New York (1972). 


3-24 Industrial Ventilation 


3.24. U.S. Public Health Service: Air Pollution Engineering 
Manual. Publication No. 999-AP-40 (1973). 


3.25. Burgess, W.A.; Ellenbecker, M.J.; Treitman, R.D.: 
Ventilation for Control of the Work Environment, 


John Wiley & Sons, New York (1989). 


3.26. Braconnier, R: Bibliographic Review of Velocity 
Fields in the Vicinity of Local Exhaust Hoods. Am. 
Ind. Hyg. Assoc. J., 49(4):185-198 (1988). 


Chapter 4 


AIR CLEANING DEVICES 


4.1. INTRODUCTION .................... 4-2 4.6.3 Thermal Oxidizers.. 2... 22.00.02... 4-25 

4.2 SELECTION OF DUST COLLECTION EQUIPMENT 4-2 4.6.4  DirectCombustors............... 4-25 
4.2.1. Contaminant Characteristics .......... 4-2 4.6.5 Catalytic Oxidizers .............. 4-25 
4.2.2 Efficiency Required. ...........2... 4-2 4.7 UNITFCOLLECTORS ...... 06.60.6400 8 +. 4-25 
4.2.3. Gas Stream Characteristics .. 2... ....0.-. 4-3 4.8 DUST COLLECTING EQUIPMENT COST ..... 4-25 
4.2.4 Contaminant Characteristics ... 2.2.0... 4-3 4.8.1 Price Versus Capacity ............. 4-25 
4.2.5 Energy Considerations ............. 4-3 4.8.2 AccessoriesIncluded ............. 4-25 
4.2.6 DustDisposal................0. 4-3 4.8.3 InstallationCost ................ 4-28 

4.3. DUST COLLECTOR TYPES .............. 4-3 4.8.4 Special Construction... ........... 4-28 
4.3.1 Electrostatic Precipitators.........2.2.. 4-3 4.9 SELECTION OF AIR FILTRATION EQUIPMENT . 4-28 
43.2 Fabric Collectors ................ 4-9 49.1 Straining 2.2... 0.20200... ...0040. 4-28 
43.3. WetCollectors ............0-.. 4-17 492 Impingement.................., 4-32 
43.4 Dry Centrifugal Collectors ........2.~. 4-18 4.9.3 Interception ........0.......00.. 4-32 

4.4 ADDITIONAL AIDS IN DUCT COLLECTOR 494 Diffusion... ......20....000.. 4.32 
SELECTION ©2222. 2 ee ee 4-22 WS” Blectrostatie. chew 2k oh ye meek 4-32 

4.5 CONTROL OF MIST, GAS, AND VAPOR 4.10 RADIOACTIVE AND HIGH TOXICITY 
CONTAMINANTS ..........0.0.00005 4-22 OPERATIONS .............- 0000025 4-33 

4.6 GASEOUS CONTAMINANT COLLECTORS .. . 4-25 4.11 EXPLOSION VENTING................ 4-33 
4.6.1 Absorbers...............0--. 4-25 REFERENCES................-- 050504. 4-34 
4.6.2 Adsorbers............2.2000% 4-25 

Figure 4-1 Dry Type Dust Collectors—Dust Disposal... . 4-4 Figure 4-1] Wet Type Dust Collectors (for Particulate 

Figure 4-2 Dry Type Dust Collectors—Discharge Valves . . 4-5 Contaminants) ..............00. 4-20 

Figure 4-3 Dry Type Dust Collectors—Discharge Valves . . 4-6 Figure 4-12. | Wet Type Dust Collector (for Particulate 

Figure 4-4 Electrostatic Precipitator, High Voltage Design . 4-7 Contaminants) ................. 4-21 

Figure 4-5 Electrostatic Precipitator, Low Voltage Design . 4-8 Figure 4-13. Dry Type Centrifugal Collectors ........ 4-23 

Figure 4-6 Performance Versus Time Between Reconditionings Figure 4-14 RangeofParticle Size... 2... ....022. 4-24 

—Fabric Collectors... ...........2.. 4-10 Figure 4-15 Unit Collector (Fabric—Shaker Type) .... . 4-29 

Figure 4-7 Fabric Collectors . 2.2... ..-2.20.0-000.4 4-14 Figure 4-16 Cost Estimates of Dust Collecting Equipment . 4-30 

Figure 4-8 Air Flow Through Fabric Collectors ... 2... 4-15 Figure 4-17 | Comparison Between Various Methods 

Figure 4-9 Fabric Collectors Pulse Jet Type... 2.2... =~. 4-16 of Measuring Air Cleaning Capability... . . 4-32 


Figure 4-10 Wet Type Collector (for Gaseous Contaminant) 4-19 


4-2 Industrial Ventilation 


4.1 INTRODUCTION 


Air cleaning devices remove contaminants from an air or 
gas stream. They are available in a wide range of designs to 
meet variations in air cleaning requirements. Degree of re- 
moval required, quantity and characteristics of the contami- 
nant to be removed, and conditions of the air or gas stream 
will all have a bearing on the device selected for any given 
application. In addition, fire safety and explosion control must 
be considered in all selections. (See NFPA publications.) 


For particulate contaminants, air cleaning devices are di- 
vided into two basic groups: AIR FILTERS and DUST COL- 
LECTORS. Air filters are designed to remove low dust 
concentrations of the magnitude found in atmospheric air. 
They are typically used in ventilation, air-conditioning, and 
heating systems where dust concentrations seldom exceed 1.0 
grains per thousand cubic feet of air and are usually well 
below 0.1 grains per thousand cubic feet of air. (One pound 
equals 7000 grains. A typical atmospheric dust concentration 
in an urban area is 87 micrograms per cubic meter or 0.038 
grains per thousand cubic feet of air.) 


Dust collectors are usually designed for the much heavier 
loads from industrial processes where the air or gas to be 
cleaned originates in local exhaust systems or process stack 
gas effluents. Contaminant concentrations will vary from less 
than 0.1 to 100 grains or more for each cubic foot of air or 
gas. Therefore, dust collectors are, and must be, capable of 
handling concentrations 100 to 20,000 times greater than 
those for which air filters are designed. 


Small, inexpensive versions ofall categories of air cleaning 
devices are available. The principles of selection, application, 
and operation are the same as for larger equipment. However, 
due to the structure of the market that focuses on small, 
quickly available, and inexpensive equipment, much of the 
available equipment is of light duty design and construction. 
One of the major economies of unit collectors implies recir- 
culation, for which such equipment may or may not be 
suitable. For adequate prevention of health hazards, fires, and 
explosions, application engineering is Just as essential for unit 
collectors as it is for major systems. 


4.2 SELECTION OF DUST COLLECTION EQUIPMENT 


Dust collection equipment is available in numerous designs 
utilizing many different principles and featuring wide vari- 
ations in effectiveness, first cost, operating and maintenance 
cost, space, arrangement, and materials of construction. Con- 
sultation with the equipment manufacturer is the recom- 
mended procedure in selecting a collector for any problem 
where extensive previous plant experience on the specific dust 
problem is not available. Factors influencing equipment se- 
lection include the following: 


4.2.1 Contaminant Characteristics: Contaminants in ex- 
haust systems cover an extreme range in concentration and 


particle size. Concentrations can range from less than 0.1 to 
much more than 100,000 grains of dust per cubic foot of air. 
In low pressure conveying systems, the dust ranges from 0.5 
to 100 or more microns in size. Deviation from mean size (the 
range over and under the mean) will also vary with the 
material. 


4.2.2 Efficiency Required: Currently, there is no accepted 
standard for testing and/or expressing the “efficiency” of a 
dust collector. It is virtually impossible to accurately compare 
the performance of two collectors by comparing efficiency 
claims. The only true measure of performance is the actual 
mass emission rate, expressed in terms such as mg/m? or 
grains/ft?. Evaluation will consider the need for high effi- 
ciency—high cost equipment requiring minimum energy such 
as high voltage electrostatic precipitators, high effi- 
ciency~moderate cost equipment such as fabric or wet collec- 
tors, or the lower cost primary units such as the dry centrifugal 
group. If either of the first two groups is selected, the combi- 
nation with primary collectors should be considered. 


When the cleaned air is to be discharged outdoors, the 
required degree of collection can depend on plant location; 
nature of contaminant (its salvage value and its potential as a 
health hazard, public nuisance, or ability to damage property); 
and the regulations of governmental agencies. In remote 
locations, damage to farms or contribution to air pollution 
problems of distant cities can influence the need for and 
importance of effective collection equipment. Many indus- 
tries, originally located away from residential areas, failed to 
anticipate the residential building construction which fre- 
quently develops around a plant. Such lack of foresight has 
required installation of air cleaning equipment at greater 
expense than initially would have been necessary. Today, the 
remotely located plant must comply, in most cases, with the 
same regulations as the plant located in an urban area. With 
the present emphasis on public nuisance, public health, and 
preservation and improvement of community air quality, 
management can continue to expect criticism for excessive 
emissions of air contaminants whether located in a heavy 
industry section of a city or in an area closer to residential 
zones. 


The mass rate of emission will also influence equipment 
selection. For a given concentration, the larger the exhaust 
volumetric flow rate, the greater the need for better equip- 
ment. Large central steam-generating stations might select 
high efficiency electrostatic precipitators or fabric collectors 
for their pulverized coal boiler stacks while a smaller indus- 
trial pulverized fuel boiler might be able to use slightly less 
efficient collectors. 


A safe recommendation in equipment selection is to select 
the collector that will allow the least possible amount of 
contaminant to escape and is reasonable in first cost and 
maintenance while meeting all prevailing air pollution regu- 
lations. For some applications even the question of reasonable 


cost and maintenance must be sacrificed to meet established 
standards for air pollution control or to prevent damage to 
health or property. 


It must be remembered that visibility of an effluent will be 
a function of the light reflecting surface area of the escaping 
material. Surface area per pound increases inversely as the 
square of particle size. This means that the removal of 80% 
or more of the dust on a weight basis may remove only the 
coarse particles without altering the stack appearance. 


4.2.3 Gas Stream Characteristics: The characteristics of 
the carrier gas stream can have a marked bearing on equip- 
ment selection. Temperature of the gas stream may limit the 
material choices in fabric collectors. Condensation of water 
vapor will cause packing and plugging of air or dust passages 
in dry collectors. Corrosive chemicals can attack fabric or 
metal in dry collectors and when mixed with water in wet 
collectors can cause extreme damage. 


4.2.4 Contaminant Characteristics: The contaminant 
characteristics will also affect equipment selection. Chemi- 
cals emitted may attack collector elements or corrode wet type 
collectors. Sticky materials, such as metallic buffing dust 
impregnated with buffing compounds, can adhere to collector 
elements, plugging collector passages. Linty materials will 
adhere to certain types of collector surfaces or elements. 
Abrasive materials in moderate to heavy concentrations will 
cause rapid wear on dry metal surfaces. Particle size, shape, 
and density will rule out certain designs. For example, the 
parachute shape of particles like the "bees wings" from grain 
will float through centrifugal collectors because their velocity 
of fall is less than the velocity of much smaller particles 
having the same specific gravity but a spherical shape. The 
combustible nature of many finely divided materials will 
require specific collector designs to assure safe operation. 


4.2.5 Energy Considerations: The cost and availability 
of energy makes essential the careful consideration of the total 
energy requirement for each collector type which can achieve 
the desired performance. An electrostatic precipitator, for 
example, might be a better selection ata significant initial cost 
penalty because of the energy savings through its inherently 
lower pressure drop. 


4.2.6 Dust Disposal: Methods of removal and disposal of 
collected materials will vary with the material, plant process, 
quantity involved, and collector design. Dry collectors can be 
unloaded continuously or in batches through dump gates, 
trickle valves, and rotary locks to conveyors or containers. 
Dry materials can create a secondary dust problem if careful 
thought is not given to dust-free material disposal or to 
collector dust bin locations suited to convenient material 
removal. See Figures 4-1, 4-2, and 4-3 for some typical 
discharge arrangements and valves. 


Wet collectors can be arranged for batch removal or con- 


Air Cleaning Devices 4-3 


tinual ejection of dewatered material. Secondary dust prob- 
lems are eliminated, although disposal of wet sludge can be a 
material handling problem. Solids carry-over in waste water 
can create a sewer or stream pollution problem if waste water 
is not properly clarified. 


Material characteristics can influence disposal problems. 
Packing and bridging of dry materials in dust hoppers, floating 
or slurry forming characteristics in wet collectors are exam- 
ples of problems that can be encountered. 


4.3 DUST COLLECTOR TYPES 


The four major types of dust collectors for particulate 
contaminants are electrostatic precipitators, fabric collectors, 
wet collectors, and dry centrifugal collectors. 


4.3.1 Electrostatic Precipitators: |n electrostatic precipi- 
tation, a high potential electric field is established between 
discharge and collecting electrodes of opposite electrical 
charge. The discharge electrode is of small cross-sectional 
area, such as a wire or a piece of flat stock, and the collection 
electrode is large in surface area such as a plate. 


The gas to be cleaned passes through an electrical field that 
develops between the electrodes. At a critical voltage, the gas 
molecules are separated into positive and negative ions. This 
is called "ionization" and takes place at, or near, the surface 
of the discharge electrode. lons having the same polarity as 
the discharge electrode attach themselves to neutral particles 
in the gas stream as they flow through the precipitator. These 
charged particles are then attracted to a collecting plate of 
opposite polarity. Upon contact with the collecting surface, 
dust particles lose their charge and then can be easily removed 
by washing, vibration, or gravity. 


The electrostatic process consists of: 
1. Tonizing the gas. 
2. Charging the dust particles. 
3. Transporting the particles to the collecting surface. 
4 


. Neutralizing, or removing the charge from the dust 
particles. 


5. Removing the dust from the collecting surface. 


The two basic types of electrostatic precipitators are "Cot- 
trell," or single-stage, and "Penny," or two-stage (see Figures 
4-4 and 4-5). 


The "Cottrell," single-stage, precipitator (Figure 4-4) com- 
bines ionization and collection in a single stage. Because it 
operates at ionization voltages from 40,000 to 70,000 volts 
DC, it may also be called a high voltage precipitator and is 
used extensively for heavy-duty applications such as utility 
boilers, larger industrial boilers, and cement kilns. Some 
precipitator designs use sophisticated voltage control systems 
and rigid electrodes instead of wires to minimize maintenance 
problems. 


4-4 Industrial Ventilation 


| 
mails 


Collector 


| 
LL 


Collector Collector 


collector sock 


Vent to collector 
or inlet duct 


x 


Covered 
tote box 
or drum 


Covered 


“Ml 


Covered drum or pail 
for dust removal 


Collector Collector 


Enclosure 


r~ Pug mill, sluice, ‘| 
x pneumatic conveyor Collapsed : 
x or screw conveyor ag 
i] pene 


Disposable bag 
or tote box 


| DRY TYPE DUST COLLECTORS | 
DUST DISPOSAL 


| AMERICAN CONFERENCE 
| OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


Air Cleaning Devices 4-5 


a Hopper 


Handle 


For intermittent manual dumping 
where dust loads are light. 


Hopper 


DUST DOOR 


Rubber gasket 


Similar to dust door but designed 
for direct attachment to dust chute, 
external pipe or canvas connection. \ 


;~ Hopper 
/ DUST GATE 


ies A Slide 
= J 
ae 2: f For intermittent, manual dumping where 


y dust loads are light. Flange for connection 
to dust disposal chute. 


SLIDE GATE 


| AMERICAN CONFERENCE | DRY TYPE DUST COLLECTORS | 
| OF GOVERNMENTAL DISCHARGE VALVES 
INDUSTRIAL HYGIENISTS k= ae , 


4-6 Industrial Ventilation 


ae Hopper 


Curtain 


For continuous removal of collected dust where hopper 
is under negative pressure. Curtain is kept closed by 
pressure differential until collected material builds up 
sufficient height to overcome pressure. 


Hopper 


Rotary valve ~ 


TRICKLE VALVE 


Drive 


Motor driven multiple blade rotary valve provide air 
lock while continuously dumping collected material. 
Can be used with hoppers under either positive or 
negative pressure. Flanged for connection to dust 

disposal chute. 


ROTARY LOCK 


Motor driven, double gate valve for continuous 
removal of collected dust. Gates are sequenced 
so only one is open ata time in order to provide 
air seal. Flanged for connection to dust disposal 
chute. 


DOUBLE DUMP VALVE 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 


| INDUSTRIAL HYGIENISTS 


DRY TYPE DUST COLLECTORS 
DISCHARGE - VALVES 


Air Cleaning Devices 4-7 


High Voltage Rectifier 
? 


~ 


Inlet Nozzle —~ 


lonizer Wires - 


Distribution Plates 


“> Wire Tensioning Weights 


Plates — 


Hoppers 


1s Collection plates 


;a seed a 


Jo colleci difficull dusts ff 
Bess Gg = Change treatment time 


i” » 1.Length 
Airflow ss 12 .Lengthen passage 
ar ; 2.Lower velocities 


access eke ag | 3.Closer plate spacing ; 
Discharge electrode gf 


C = Se Sa 4} 


First field 


Second field 


ELECTROSTATIC PRECIPITATOR | 
HIGH VOLTAGE DESIGN 
(40,000 TO 75,000 VOLTS) 


oe Tae . : Trou s ee ee 


| AMERICAN CONFERENCE 
f OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


4-8 Industrial Ventilation 


Trash screen 
and distribution 
baffle 


Side access door 


Power pack 
i, 
H oO Fan section 
4 m 

fhA 
— 
< 
< 


XK XX XK 


XXKX XS 


Air flow 


ica eee eg! 


EK 


Spray nozzle 
header 


lonizer 
wire 


™ Insulator 
Plates 


Grounded plates 


Charged plates 


Discharge electrode 


Collection plates 
(Grounded) 


ELECTROSTATIC PRECIPITATOR | 
Seen LOW VOLTAGE DESIGN 
p (OF = GOVERNMENTAL (11,000 TO 15,000 VOLTS) 
rcs Sa a A FA 


ICAN CONFERENCE | 


| AMER 


The "Penny," or two-stage, precipitator (Figure 4-5) uses 
DC voltages from 11,000 to 14,000 volts for ionization and 
is frequently referred to as a low voltage precipitator. Its use 
is limited to low concentrations, normally not exceeding 
0.025 grains per cubic foot. It is the most practical collection 
technique for the many hydrocarbon applications where an 
initially clear exhaust stack turns into a visible emission as 
vapor condenses. Some applications include plasticizer ov- 
ens, forge presses, die-casting machines, and various welding 
operations. Care must be taken to keep the precipitator inlet 
temperature low enough to insure that condensation has al- 
ready occurred. 


For proper results, the inlet gas stream should be evaluated 
and treated where necessary to provide proper conditions for 
ionization. For high voltage units, a cooling tower is some- 
times necessary. Low voltage units may use wet scrubbers, 
evaporative coolers, heat exchangers or other devices to con- 
dition the gas stream for best precipitator performance. 


The pressure drop of an electrostatic precipitator is ex- 
tremely low, usually Jess than | "wg; therefore, the energy 
requirement is significantly less than for other techniques. 


4.3.2 Fabric Collectors: Fabric collectors remove par- 
ticulate by straining, impingement, interception, diffusion, 
and electrostatic charge. The "fabric" may be constructed of 
any fibrous material, either natural or man-made, and may be 
spun into a yarn and woven or felted by needling, impacting, 
or bonding. Woven fabrics are identified by thread count and 
weight of fabric per unit area. Non-woven (felts) are identified 
by thickness and weight per unit area. Regardless of construc- 
tion, the fabric represents a porous mass through which the 
gas is passed unidirectionally such that dust particles are 
retained on the dirty side and the cleaned gas passes on 
through. 


The ability of the fabric to pass air is stated as "permeabil- 
ity" and is defined as the cubic feet of air passed through one 
square foot of fabric each minute at a pressure drop of 0.5 
“we. Typical permeability values for commonly used fabrics 
range from 25 to 40 cfm. 


A non-woven (felted) fabric is more efficient than a woven 
fabric of identical weight because the void areas or pores in 
the non-woven fabric are smaller. A specific type of fabric 
can be made more efficient by using smaller fiber diameters, 
a greater weight of fiber per unit area, and by packing the 
fibers more tightly. For non-woven construction, the use of 
finer needles for felting also improves efficiency. While any 
fabric is made more efficient by these methods, the cleanabil- 
ity and permeability are reduced. A highly efficient fabric that 
cannot be cleaned represents an excessive resistance to air 
flow and is not an economical engineering solution. Final 
fabric selection is generally a compromise between efficiency 
and permeability. 


Choosing a fabric with better cleanability or greater perme- 


Air Cleaning Devices 4-9 


ability but lower inherent efficiency is not as detrimental as 
it may seem. The efficiency of the fabric as a filter is mean- 
ingful only when new fabric is first put into service. Once the 
fabric has been in service any length of time, collected par- 
ticulate in contact with the fabric acts as a filter aid, improving 
collection efficiency. Depending on the amount of particulate 
and the time interval between fabric reconditioning, it may 
well be that virtually all filtration is accomplished by the 
previously collected particulate—or dust cake~-as opposed 
to the fabric itself. Even immediately after cleaning, a residual 
and/or redeposited dust cake provides additional filtration 
surface and higher collection efficiency than obtainable with 
new fabric. While the collection efficiency of new, clean 
fabric is easily determined by laboratory test and the informa- 
tion is often published, it is not representative of operating 
conditions and therefore is of little importance in selecting the 
proper collector. 


Fabric collectors are not 100% efficient, but well-designed, 
adequately sized, and properly operated fabric collectors can 
be expected to operate at efficiencies in excess of 99%, and 
often as high as 99.9% or more on a mass basis. The ineffi- 
ciency, or penetration, that does occur is greatest during or 
immediately after reconditioning. Fabric collector ineffi- 
ciency is frequently a result of by-pass due to damaged fabric, 
faulty seals, or sheet metal leaks rather than penetration of the 
fabric. Where extremely high collection efficiency is essen- 
tial, the fabric collector should be leak tested for mechanical 
leaks. 


The combination of fabric and collected dust becomes 
increasingly efficient as the dust cake accumulates on the 
fabric surface. At the same time, the resistance to air flow 
increases. Unless the air moving device is adjusted to com- 
pensate for the increased resistance, the gas flow rate will be 
reduced. Figure 4-6 shows how efficiency, resistance to flow, 
and flow rate change with time as dust accumulates on the 
fabric. Fabric collectors are suitable for service on relatively 
heavy dust concentrations. The amount of dust collected ona 
single square yard of fabric may exceed five pounds per hour. 
In virtually all applications, the amount of dust cake accumu- 
lated in just a few hours will represent sufficient resistance to 
flow to cause an unacceptable reduction in air flow. 


In a well-designed fabric collector system, the fabric or 
filter mat is cleaned or reconditioned before the reduction in 
air flow is critical. The cleaning is accomplished by mechani- 
cal agitation or air motion, which frees the excess accumula- 
tion of dust from the fabric surface and leaves a residual or 
base cake. The residual dust cake does not have the same 
characteristics of efficiency or resistance to air flow as new 
fabric. 


Commercially available fabric collectors employ fabric 
configured as bags or tubes, envelopes (flat bags), rigid 
elements, or pleated cartridges. Most of the available fabrics, 
whether woven or non-woven, are employed in either bag or 


4-10 Industrial Ventilation 


Collection efficiency 


Collection efficiency, percent by weight 
Pressure drop, inches water gauge 
Air flow, cubic feet per minute 


Time, from last reconditioning 


ae eT 
BETWEEN RECONDITIONINGS — | 
FABRIC_COLLECTORS | 


| AMERICAN CONFERENCE 
| OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


envelope configuration. The pleated cartridge arrangement 
uses a paper-like fiber in either a cylindrical or panel configu- 
ration. It features extremely high efficiency on light concen- 
trations. Earlier designs employed cellulose-based media. 
Today, more conventional media, such as polypropolene or 
spun-bonded polyester, are frequently used. 


The variable design features of the many fabric collectors 
available are: 


|. Type of fabric (woven or non-woven). 


2. Fabric configuration (bags or tubes, envelopes, car- 
tridges). 


3. Intermittent or continuous service. 
4. Type of reconditioning (shaker, pulse-jet, reverse-air). 


5. Housing configuration (single compartment, multiple 
compartment). 


At least two of these features will be interdependent. For 
example, non-woven fabrics are more difficult to recondition 
and therefore require high-pressure cleaning. 


A fabric collector is selected for its mechanical, chemical, 
and thermal characteristics. Table 4-1 lists those charac- 
teristics for some common filter fabrics. 


Fabric collectors are sized to provide a sufficient area of 
filter media to allow operation without excessive pressure 
drop. The amount of filter area required depends on many 
factors, including: 


1. Release characteristics of dust. 

. Porosity of dust cake. 

. Concentration of dust in carrier gas stream. 
. Type of fabric and surface finish, if any. 

. Type of reconditioning. 


. Reconditioning interval. 


NID NW HP WwW HY 


. Air flow pattern within the collector. 
8. Temperature and humidity of gas stream. 


Because of the many variables and their range of variation, 
fabric collector sizing is a judgment based on experience. The 
sizing is usually made by the equipment manufacturer, but at 
times may be specified by the user or a third party. Where no 
experience exists, a pilot installation is the only reliable way 
to determine proper size. 


The sizing or rating of a fabric collector is expressed in 
terms of air flow rate versus fabric media area. The resultant 
ratio is called "air to cloth ratio" with units of cfm per square 
foot of fabric. This ratio represents the average velocity of the 
gas stream through the filter media. The expression "filtration 
velocity" is used synonymously with air to cloth ratio for 
rating fabric collectors. For example, an air to cloth ratio of 
7:1 (7 cfm/sq ft) is equivalent to a filtration velocity of 7 fpm. 


Table 4-2 compares the various characteristics of fabric 


Air Cleaning Devices 4-1] 


collectors. The different types will be described in detail later. 
Inspection of Table 4-2 now may make the subsequent dis- 
cussion more meaningful. The first major classification of 
fabric collectors is intermittent or continuous duty. Intermit- 
tent-duty fabric collectors cannot be reconditioned while in 
operation. By design, they require that the gas flow be inter- 
rupted while the fabric is agitated to free accumulated dust 
cake. Continuous-duty collectors do not require shut down for 
reconditioning. 


Intermittent-duty fabric collectors may use a tube, car- 
tridge, or envelope configuration of woven fabric and will 
generally employ shaking or vibration for reconditioning. 
Figure 4-7 shows both tube and envelope shaker collector 
designs. For the tube type, dirty air enters the open bottom of 
the tube and dust is collected on the inside of the fabric. The 
bottoms of the tubes are attached to a tube sheet and the tops 
are connected to a shaker mechanism. Since the gas flow is 
from inside to outside, the tubes tend to inflate during opera- 
tion and no other support of the fabric is required. 


Gas flow for envelope-type collectors is from outside to 
inside; therefore, the envelopes must be supported during 
operation to prevent collapsing. This is normally done by 
inserting wire mesh or fabricated wire cages into the enve- 
lopes. The opening of the envelope from which the cleaned 
air exits is attached to a tube sheet and, depending on design, 
the other end may be attached to a support member or canti- 
levered without support. The shaker mechanism may be lo- 
cated in either the dirty air or cleaned air compartments. 


Periodically (usually at 3- to 6-hour intervals) the air flow 
must be stopped to recondition the fabric. Figure 4-8 illus- 
trates the system air flow characteristics of an intermittent- 
duty fabric collector. As dust accumulates on the fabric, 
resistance to flow increases and air flow decreases until the 
fan is turned off and the fabric reconditioned. Variations in 
air flow due to changing pressure losses is sometimes a 
disadvantage and, when coupled with the requirement to 
periodically stop the air flow, may preclude the use of inter- 
mittent collectors. Reconditioning seldom requires more than 
two minutes but must be done without air flow through the 
fabric. If reconditioning is attempted with air flowing, it will 
be less effective and the flexing of the woven fabric will allow 
a substantial amount of dust to escape to the clean air side. 


The filtration velocity for large intermittent-duty fabric 
collectors seldom exceeds 6 fpm and normal selections are in 
the 2-4 fpm range. Lighter dust concentrations and the ability 
to recondition more often allow the use of higher filtration 
velocities. Ratings are usually selected so that the pressure 
drop across the fabric will be in the 2—5 "wg range between 
start and end of operating cycle. 


With multiple-section, continuous-duty, automatic fabric 
collectors, the disadvantage of stopping the air flow to permit 
fabric reconditioning and the variations in air flow with dust 
cake build-up can be overcome. The use of sections or com- 


TABLE 4—1. Characteristics of Filter Fabrics* 


tiv 


Example Max. Temp. F Resistance to Physical Action Resistance to Chemicals 
Generic Trade Name 
Names Fabrics™ Continuous Intermittent Dry Heat Moist Heat Abrasion Shaking Flexing Mineral Acid Organic Acid Alkalies Oxidizing Solvents 
Cotton Cotton 180 — G G F G G P G F F E 
Polyester Dacron 
Fortrel®) 
Vycron®) 
Kodel*) 
Enka 
Polyester’) 275 — G F G E E G G F G E 
Acrylic Orlon) 
Acrilan® 
Creslan” 
Dralon T‘®) 
Zefran 275 285 G G G G E G G F G E 
Modacrylic Dynel(9) 
Verel(®) 160 _ F F F P-F G G G G G G 
Nylon Nylon 
(Polyamide) 6,6(126) 
Nylon 66:11:12) 225 _ G G E E E P F G F E 
Nomext") 400 450 E E E E E P-F E G G E 
Polymide P-84(18) 500 580 E P G G E P-F G F G E 
Polypropylene Herculon''9) 
Reevon'"4) 
Vectra('®) 200 250 G F E E G E E E G G 
Teflon Teflon 
(flurocarbon) TFE) 500 550 E E P-F G G E E E E E 
Teflon 
FEPU) 450 — E E P-F G G E & E E E 
Expanded Rastex 500 550 E E P-F G G E E E E E 
PFTE 
Vinyon Vinyon("®) 
Clevyit(™” 350 — F F F G G E E G G P 
Glass Glass 500 600 E E P P F E E F E E 
Fiberglass _Fiberglass‘'®) 550 550 E E P P G G G G E G 


*E - excellent; G = good; F = fair; P = poor 
* Registered Trademarks 


(1) Du Pont; (2) Celanese; (3) Beaunit; (4) Eastman; (5) American Enka; (6) Chemstrand; (7) American Cyanamid; (8) Farbenfabriken Bayer AG; (9) Dow Chemical; (10) Union Carbide; (11) Allied Chemical; (12) Firestone; (13) Hercules; 
(14) Alamo Polymer; (15) National Plastic; (16) FMC; (17) Societe Rhovyl; (18) Lenzing; (19) Huyglas 


vores QUI, [eLysnpuy 


partments, as indicated in Figure 4-7, allows continuous 
operation of the exhaust system because automatic dampers 
periodically remove one section from service for fabric recon- 
ditioning while the remaining compartments handle the total 
gas flow. The larger the number of compartments, the more 
constant the pressure loss and air flow. Either tubes or enve- 
lopes may be used and fabric reconditioning is usually accom- 
plished by shaking or vibrating. 


Figure 4-8 shows air flow versus time for a multiple-section 
collector. Each individual section or compartment has an air 
flow versus time characteristic like that of the intermittent 
collector, but the total variation is reduced because of the 
multiple compartments. Note the more constant air flow 
characteristic of the five-compartment unit as opposed to the 
three-compartment design. Since an individual section is out 
of service only a few minutes for reconditioning and remain- 
ing sections handle the total gas flow during that time, it is 
possible to clean the fabric more frequently than with the 
intermittent type. This permits the multiple-section unit to 
handle higher dust concentrations. Compartments are recon- 
ditioned in fixed sequence with the ability to adjust the time 
interval between cleaning of individual compartments. 


One variation of this design is the low-pressure, reverse-air 
collector which does not use shaking for fabric recondition- 
ing. Instead, a compartment is isolated for cleaning and the 
tubes collapsed by means ofa low pressure secondary blower, 
which draws air from the compartment in a direction opposite 
to the primary air flow. This is a "gentle" method of fabric 
reconditioning and was developed primarily for the fragile 
glass cloth used for high-temperature operation. The reversal 
of air flow and tube deflation is accomplished very gently to 
avoid damage to the glass fibers. The control sequence usually 
allows the deflation and re-inflation of tubes several times for 
complete removal of excess dust. Tubes are 6-11 inches in 
diameter and can be as long as 30 feet. For long tubes, stainless 
steel rings may be sewn on the inside to help break up the dust 


Table 4-2. Summary of Fabric-Type Collectors and Their Characteristics 


Air Cleaning Devices 4-13 


cake during deflation. A combination of shaking and reverse 
air flow has also been utilized. 


When shaking Is used for fabric reconditioning, the filtra- 
tion velocity usually is in the 1-4 fpm range. Reverse-air 
collapse-type reconditioning generally necessitates lower fil- 
tration velocities since reconditioning is not as complete. 
They are seldom rated higher than 3 fpm. The air to cloth ratio 
or filtration velocity is based on net cloth area available when 
a compartment is out of service for reconditioning. 


Reverse-jet, continuous-duty fabric collectors may use en- 
velopes or tubes of non-woven (felted) fabric, pleated car- 
tridges of non-woven mat (paper-like) in cylindrical or panel 
configuration, or rigid elements such as sintered polyethyl- 
ene. They differ from the low-pressure reverse-air type in that 
they employ a brief burst of high-pressure air to recondition 
the fabric. Woven fabric is not used because it allows exces- 
sive dust penetration during reconditioning. The most com- 
mon designs use compressed air at 80-100 psig, while others 
use an integral pressure blower at a lower pressure but higher 
secondary flow rate. Those using compressed air are generally 
called pulse-jet collectors and those using pressure blowers 
are called fan-pulse collectors. 


All designs collect dust on the outside and have air flow 
from outside to inside the fabric. All recondition the media 
by introducing the pulse of cleaning air into the opening where 
cleaned air exits from the tube, envelope, or cartridge. In many 
cases, a venturi shaped fitting is used at this opening to 
provide additional cleaning by inducing additional air flow. 
The venturi also directs or focuses the cleaning pulse for 
maximum efficiency. 


Figure 4-9 shows a typical pulse-jet collector. Under nor- 
mal operation (air flow from outside to inside), the fabric 
shape will tend to collapse; therefore, a support cage is 
required. The injection of a short pulse of high-pressure air 
induces a secondary flow from the clean air compartment in 


INTERRUPTABLE OPERATION INTERRUPTABLE OPERATION CONTINUOUS OPERATIONS 
Light to Moderate Loading Heavy Loading Any Loading 
Fabric Reconditioning 
Requirement Intermittent Continuous 
Type of Reconditioning Shaker Reverse Air Reverse Pulse — (High Pressure) 


Collector Configuration Single Compartment 


Shaker 
(Low Pressure) 


Pulse Jet of Fan Pulse 


Multiple Compartment 
with inlet or outlet dampers for each 


Single Compartment 


Fabric Configuration Tube, Cartridge or Envelope Tube or Envelope | Tube Tube or Envelope Pleated Cartridge 
Type of Fabric Woven Woven Non-Woven (Felt) Non-Woven 

Air Flow Highly Variable Slightly Variable Virtually Constant Virtually Constant | 
Normal Rating 1 to 6 fpm <1 to 7 fom 


(filtration velocity, fpm) | 


1 to 3 fpm | 1 to 3 fpm 5 to 12 fpm 


4-14 Industrial Ventilation 


as — Motor driven vibrator 

y, ~ Boffle 
a soa iho T atte nese 

ik “~ Clean 


air 


~~ outlet 


Clean Jf i 
Dusty air e 


a oir ou'let lea! 
as inlet | 
| 
Motor driven at -_ Dust outlet 
vibrator 


ENVELOPE TYPE 


ae ee fl i ii ee ce 


i Tran 


(exe) ere) 
ee 


— -— 


= : 

: : “Reverse air flow 

Screen rapping Clean air side AS Reve 
mechanism { 


i Three position outlet vaives 
Compartmenis 1,2, and 3 Sa 7 


under air-foad. Compartment stata! ; 


4 closed off for fabric 
cleaning. 


MULTIPLE SECTION CONTINUOUS AUTOMATIC 


| AMERICAN CONFERENCE | 
| OF GOVERNMENTAL — [| FABRIC Ree ears 


Air Cleaning Devices 4-15 


/ 


AIR FLOW, cfm 
Reconditionin 


“2 


One cycle 
TIME 


INTERMITTENT DUTY FABRIC COLLECTOR 


Reconditioning 
Reconditioning 
i 
Reconditioning 
ran 


/ Reconditioning 


Reconditioning 
Reconditioning 
Reconditioning 


AIR FLOW, cfm 
Ss 


— 


SS One cycles ‘ ee One cycle ——— = i 
TIME TIME 


3 Compartment 5 Compartment 


MULTIPLE SECTION, CONTINUOUS DUTY FABRIC COLLECTOR 


Note: 
The flow variation has been exaggerated. 


AlR FLOW THROUGH 
PADRIC. COLLECTORS 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


4-16 Industrial Ventilation 


Dirty air inlet ee s Clean air outlel 


eee Fiber envelope 


Reverse air 
jet nozzles 


a 


Rotary valve 


Dust ouflet 


Collection Pail 


Reverse jet piping 


Clean air outlet “X. 
1 


NX 


Solenoid valves & controls 


— Fabric element 


| 


Dirty air intet a 
AG 


| AMERICAN CONFERENCE | FABRIC COLLECTORS 
OF GOVERNMENTAL | PULSE JET TYPE | 


a direction opposite to the normal air flow. Reconditioning is 
accomplished by the pulse of high-pressure air which stops 
forward air flow, then rapidly pressurizes the media, breaking 
up the dust cake and freeing accumulated dust from the fabric. 
The secondary or induced air acts as a damper, preventing 
flow in the normal direction during reconditioning. The entire 
process, from injection of the high-pressure pulse and initia- 
tion of secondary flow until the secondary flow ends, takes 
place in approximately one second. Solenoid valves which 
control the pulses of compressed air may be open for a tenth 
of a second or less. An adequate flow rate of clean and dry 
compressed air of sufficient pressure must be supplied to 
ensure effective reconditioning. 


Reverse-jet collectors normally clean no more than 10% of 
the fabric at any one time. Because such a small percentage 
is cleaned at any one time and because the induced secondary 
flow blocks normal flow during that time, reconditioning can 
take place while the collector is in service and without the 
need for compartmentation and dampers. The cleaning inter- 
vals are adjustable and are considerably more frequent than 
the intervals for shaker or reverse-air collectors. An individual 
element may be pulsed and reconditioned as often as once a 
minute to every six minutes. 


Due to this very short reconditioning cycle, higher filtration 
velocities are possible with reverse-jet collectors. However, 
with all reverse-jet collectors, accumulated dust that is freed 
from one fabric surface may become reintrained and redepo- 
sited on an adjacent surface, or even on the original surface. 
This phenomenon of redeposition tends to limit filtration 
velocity to something less than might be anticipated with 
cleaning intervals of just a few minutes. 


Laboratory tests“ have shown that for a given collector 
design redeposition increases with filtration velocity. Other 
test work“ indicates clearly that redeposition varies with 
collector design and especially with flow patterns in the dirty 
air compartment. EPA-sponsored research“) has shown that 
superior performance results from downward flow of the dirty 
air stream. This downward air flow reduces redeposition since 
it aids gravity in moving dust particles toward the hopper. 


Filtration velocities of S—12 fpm are normal for reverse-jet 
collectors. The pleated cartridge type of reverse-jet collector 
is limited to filtration velocities in the 7 fpm range. The pleat 
configuration may produce very high approach velocities and 
greater redeposition. 


4.3.3 Wet Collectors: Wet collectors, or scrubbers, are 
commercially available in many different designs, with pres- 
sure drops from 1.5 "wg to as much as 100 "wg. There is a 
corresponding variation in collector performance. It is gener- 
ally accepted that, for well-designed equipment, efficiency 
depends on the energy utilized in air to water contact and is 
independent of operating principle. Efficiency is a function 
of total energy input per cfm whether the energy is supplied 


Air Cleaning Devices 4-17 


to the air or to the water. This means that well-designed 
collectors by different manufacturers will provide similar 
efficiency if equivalent power is utilized. 


Wet collectors have the ability to handle high-temperature 
and moisture-laden gases. The collection of dust in a wetted 
form minimizes a secondary dust problem in disposal of 
collected material. Some dusts represent explosion or fire 
hazards when dry. Wet collection minimizes the hazard; 
however, the use of water may introduce corrosive conditions 
within the collector and freeze protection may be necessary 
if collectors are located outdoors in cold climates. Space 
requirements are nominal. Pressure losses and collection ef- 
ficiency vary widely for different designs. 


Wet collectors, especially the high-energy types, are fre- 
quently the solution to air pollution problems. It should be 
recognized that disposal of collected material in water without 
clarification or treatment may create water pollution prob- 
lems. 


Wet collectors have one characteristic not found in other 
collectors — the inherent ability to humidify. Humidification, 
the process of adding water vapor to the air stream through 
evaporation, may be either advantageous or disadvantageous 
depending on the situation. Where the initial air stream is at 
an elevated temperature and not saturated, the process of 
evaporation reduces the temperature and the volumetric flow 
rate of the gas stream leaving the collector. Assuming the fan 
is to be selected for operation on the clean air side of the 
collector, it may be smaller and will definitely require less 
power than if there had been no cooling through the collector. 
This is one of the obvious advantages of humidification; 
however, there are other applications where the addition of 
moisture to the gas stream is undesirable. For example, the 
exhaust of humid air to an air-conditioned space normally 
places an unacceptable load on the air conditioning system. 
High humidity can also result in corrosion of finished goods. 
Therefore, humidification effects should be considered before 
designs are finalized. While all wet collectors humidify, the 
amount of humidification varies for different designs. Most 
manufacturers publish the humidifying efficiency for their 
equipment and will assist in evaluating the results. 


Chamber or Spray Tower: Chamber or spray tower collec- 
tors consist of a round or rectangular chamber into which 
water is introduced by spray nozzles. There are many vari- 
ations of design, but the principal mechanism is impaction of 
dust particles on the liquid droplets created by the nozzles. 
These droplets are separated from the air stream by centrifugal 
force or impingement on water eliminators. 


The pressure drop is relatively low (on the order of 0.5—1.5 
"we), but water pressures range from 10-400 psig. The high 
pressure devices are the exception rather than the rule. In 
general, this type of collector utilizes low-pressure supply 
water and operates in the lower efficiency range for wet 


4-18 Industrial Ventilation 


collectors. Where water is supplied under high pressure, as 
with fog towers, collection efficiency can reach the upper 
range of wet collector performance. 


For conventional equipment, water requirements are rea- 
sonable, with a maximum of about 5 gpm per thousand scfm 
of gas. Fogging types using high water pressure may require 
as much as 10 gpm per thousand scfm of gas. 


Packed Towers: Packed towers (see Figure 4-10) are es- 
sentially contact beds through which gases and liquid pass 
concurrently, counter-currently, or in cross-flow. They are 
used primarily for applications involving gas, vapor, and mist 
removal. These collectors can capture solid particulate matter, 
but they are not used for that purpose because dust plugs the 
packing and requires unreasonable maintenance. 


Water rates of 5—10 gpm per thousand scfm are typical for 
packed towers. Water is distributed over V-notched ceramic 
or plastic weirs. High temperature deterioration is avoided by 
using brick linings, allowing gas temperatures as high as 1600 
F to be handled direct from furnace flues. 


The air flow pressure loss for a four foot bed of packing, 
such as ceramic saddles, will range from 1.5-3.5 "wg. The 
face velocity (velocity at which the gas enters the bed) will 
typically be 200-300 fpm. 


Wet Centrifugal Collectors: Wet centrifugal collectors (see 
Figure 4-11) comprise a large portion of the commercially 
available wet collector designs. This type utilizes centrifugal 
force to accelerate the dust particle and impinge it upon a 
wetted collector surface. Water rates are usually 2—5 gpm per 
thousand scfm of gas cleaned. Water distribution can be from 
nozzles, gravity flow or induced water pickup. Pressure drop 
is in the 2-6 "wg range. 


As a group, these collectors are more efficient than the 
chamber type. Some are available with a variable number of 
impingement sections. A reduction in the number of sections 
results in lower efficiency, lower cost, less pressure drop, and 
smaller space. Other designs contain multiple collecting 
tubes. For a given air flow rate, a decrease in the tube size 
provides higher efficiency because the centrifugal force is 
greater. 


Wet Dynamic Precipitator: The wet dynamic precipitator 
(see Figure 4-12) is a combination fan and dust collector. Dust 
particles in the dirty air stream impinge upon rotating fan 
blades wetted with spray nozzles. The dust particles impinge 
into water droplets and are trapped along with the water by a 
metal cone while the cleaned air makes a turn of 180 degrees 
and escapes from the front of the specially shaped impeller 
blades. Dirty water from the water cone goes to the water and 
sludge outlet and the cleaned air goes to an outlet section 
containing a water elimination device. 


Orifice Type: In this group of wet collector designs (see 


Figure 4-12), the air flow through the collector is brought in 
contact with a sheet of water in a restricted passage. Water 
flow may be induced by the velocity of the air stream or 
maintained by pumps and weirs. Pressure losses vary from | 
"we or less for a water wash paint booth to a range of 3-6 "wg 
for most of the industrial designs. Pressure drops as high as 
20 "wg are used with some designs intended to collect very 
small particles. 


Venturi: The venturi collector (see Figure 4-11) uses a 
venturi-shaped constriction to establish throat velocities con- 
siderably higher than those used by the orifice type. Gas 
velocities through venturi throats may range from 
12,000-24,000 fpm. Water is supplied by piping or jets at or 
ahead of the throat at rates from 5~15 gpm per thousand scfm 
of gas. 


The collection mechanism of the venturi is impaction. As 
is true for all well-designed wet collectors, collection effi- 
ciency increases with higher pressure drops. Specific pressure 
drops are obtained by designing for selected velocities in the 
throat. Some venturi collectors are made with adjustable 
throats allowing operation over a range of pressure drops for 
a given flow rate or over a range of flow rates with a constant 
pressure drop. Systems are available with pressure drops as 
low as 5 "wg for moderate collection efficiency and as high 
as 100 "wg for collection of extremely fine particles. 


The venturi itself is a gas conditioner causing intimate 
contact between the particulates in the gas and the multiple 
jet streams of scrubbing water. The resulting mixture of gases, 
fume-dust agglomerates, and dirty water must be channeled 
through a separation section for the elimination of entrained 
droplets as shown in Figure 4-11. 


4.3.4 Dry Centrifugal Collectors: Dry centrifugal collec- 
tors separate entrained particulate from an air stream by the 
use or combination of centrifugal, inertial, and gravitational 
force. Collection efficiency is influenced by: 


1. Particle size, weight and shape. Performance is im- 
proved as size and weight become larger and as the 
shape becomes more spherical. 


2. Collector size and design. The collection of fine dust 
with a mechanical device requires equipment designed 
to best utilize mechanical forces and fit specific appli- 
cation needs. 


3. Velocity. Pressure drop through a cyclone collector 
increases approximately as the square of the inlet 
velocity. There is, however, an optimum velocity that 
is a function of collector design, dust characteristics, 
gas temperature and density. 


4. Dust concentration. Generally, the performance of a 
mechanical collector increases as the concentration of 
dust becomes greater. 


Air Cleaning Devices 4-19 


i Water in 


Distributor 


Outlet 


= 


{ 


ON 


SK 
SS 
5S 
Se 


Oa 
Se 
<S 
Sy 


ie 


SS 


Suitable packing 


media ——~ 


Steel cylindrical 
ye jacket 


Sa 


—~ Corrosion linin 
4 | an where require 


Se 
SS 
OSs 


aa «Support plate 


[}- 


Air intet 


Water Quflet 


PACKED TOWER 


WEE, TYPE COLLECTOR 
(FOR GASEOUS CONTAMINANT) | 


pate 7-88 (=?) fricurE 4-70 A= : 


ICAN CONFERENCE | 
OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


| AMER 


4-20 Industrial Ventilation 


Symbols Parts 


Clean air outlet. 
| Entrainment separator. 

Water inlet. 

Impingement plates. 

Dirty air inlet. 

| Wet cyclone for collecting heavy 
; material. 

| Water and sludge drain. 


aMmMoOoOY 


WET CENTRIFUGAL 


\ OA 


Venturi 


Se 


VENTURI SCRUBBER 


WET TYPE DUST COLLECTORS. | 
(FOR PARTICULATE 
CONTAMINANTS) 


| AMERICAN CONFERENCE | 
| OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


Air Cleaning Devices 4-21 


Entrainment 
separators —_, 


COLLECTIN 


Dirt and water 
discharged at 
blade tips. 


TYPICAL WET 
ORIFICE TYPE COLLECTOR 


Dirty air 
inlet. 


Water spray nozzle. 


Clean air outlet. 


Water and 
sludge ouffet. 


: ees vee re Vee (FOR PARTICULATE 
CONTAMINANTS) ___ 


| INDUSTRIAL HYGIENISTS Par_—j-gg_ ee __ a7 | 


4-22 Industrial Ventilation 


Gravity Separators: Gravity separators consist of a cham- 
ber or housing in which the velocity of the gas stream is made 
to drop rapidly so that dust particles settle out by gravity. 
Extreme space requirements and the usual presence of eddy 
currents nullify this method for removal of anything but 
extremely coarse particles. 


Inertial Separators: Inertial separators depend on the in- 
ability of dust to make a sharp turn because its inertia is much 
higher than that of the carrier gas stream. Blades or louvers 
in a variety of shapes are used to require abrupt turns of 120° 
or more. Well-designed inertial separators can separate parti- 
cles in the 10-20 micron range with about 90% efficiency. 


Cyclone Collector: The cyclone collector (see Figure 4-13) 
is commonly used for the removal of coarse dust from an air 
stream, as aprecleaner to more efficient dust collectors, and/or 
as a product separator in air conveying systems. Principal 
advantages are low cost, low maintenance, and relatively low 
pressure drops (in the 0.75—1.5 "wg range). It is not suitable 
for the collection of fine particles. 


High Efficiency Centrifugals: High-efficiency centrifugals 
(see Figure 4-13) exert higher centrifugal forces on the dust 
particles in a gas stream. Because centrifugal force is a 
function of peripheral velocity and angular acceleration, im- 
proved dust separation efficiency has been obtained by: 


I. Increasing the inlet velocity. 

2. Making the cyclone body and cone longer. 

3. Using a number of small diameter cyclones in parallel. 
4. Placing units in series. 


While high-efficiency centrifugals are not as efficient on 
small particles as electrostatic, fabric, and wet collectors, their 
effective collection range is appreciably extended beyond that 
of other mechanical devices. Pressure losses of collectors in 
this group range from 3-8 "wg. 


4.4 ADDITIONAL AIDS IN DUST COLLECTOR SELECTION 


The collection efficiencies of the five basic groups of air 
cleaning devices have been plotted against mass mean particle 
size (Figure 4-14). The graphs were found through laboratory 
and field testing and were not compiled mathematically. The 
number of lines for each group indicates the range that can be 
expected for the different collectors operating under the same 
principle. Variables, such as type of dust, velocity of air, water 
rate, etc., will also influence the range for a particular appli- 
cation. 


Deviation lines shown in the upper right hand corner of the 
chart allow the estimation of mass mean material size in the 
effluent of a collector when the inlet mean size is known. 
Space does not permit a detailed explanation of how the slopes 
of these lines were determined, but the following example 
illustrates how they are used. The deviation lines should not 


be used for electrostatic precipitators but can be used for the 
other groups shown at the bottom of the figure. 


Example: A suitable collector will be selected for a lime 
kiln to illustrate the use of the chart. Referring to Figure 4-14, 
the concentration and mean particle size of the material leav- 
ing the kiln can vary between 3 and 10 grains per cubic foot, 
with 5-10 microns the range for mass mean particle size. 
Assume an inlet concentration of 7.5 grains per cubic foot and 
a mean inlet size of 9 microns. Projection of this point 
vertically downwardly to the collection efficiency portion of 
the chart will indicate that a low-resistance cyclone will be 
less than 50% efficient; a high-efficiency centrifugal will be 
60-80% efficient and a wet collector, fabric arrester and 
electrostatic precipitator will be 97% efficient or more. A 
precleaner is usually feasible for dust concentrations over 5 
grains per cubic foot unless it is undesirable to have the 
collected dust separated by size. For this example a high-ef- 
ficiency centrifugal will be selected as the precleaner. The 
average efficiency is 70% for this group, therefore the effluent 
from this collector will have a concentration of 7.5 (1.00 — 
0.70) = 2.25 grains per cubic foot. Draw a line through the 
initial point with a slope parallel to the deviation lines marked 
"industrial dust." Where deviation is not known, the average 
of this group of lines normally will be sufficiently accurate to 
predict the mean particle size in the collector effluent. A 
vertical line from the point of intersection between the 2.25 
grains per cubic foot horizontal and the deviation line to the 
base of the chart will indicate a mean effluent particle size of 
6.0 microns. 


A second high-efficiency centrifugal in series would be less 
than 50% efficient on this effluent. A wet collector, fabric 
arrester, or electrostatic would have an efficiency of 94% or 
better. Assume that a good wet collector will be 98% efficient. 
The effluent would then be 2.25 (1.00-0.98) =0.045 grains 
per cubic foot. Using the previous deviation line and its 
horizontal intersection of 0.045 grains per cubic foot yields a 
vertical line intersecting the mean particle size chart at 1.6 
microns, the mean particle size of the wet collector effluent. 


In Table 4-3, an effort has been made to report types of dust 
collectors used for a wide range of industrial processes. While 
many of the listings are purely arbitrary, they may serve as a 
guide in selecting the type of dust collector most frequently 
used. 


4.5 CONTROL OF MIST, GAS, AND VAPOR 
CONTAMINANTS 


Previous discussion has centered on the collection of dust 
and fume or particulate existing in the solid state. Only the 
packed tower was singled out as being used primarily to collect 
mist, gas, or vapor. The character ofa mist aerosol is very similar, 
aerodynamically, to that of a dust or fume aerosol, and the mist 
can be removed from an air stream by applying the principles 
that are used to remove solid particulate. 


Air Cleaning Devices 4-23 


/* 


LOW PRESSURE CYCLONE 


HIGH EFFICIENCY CENTRIFUGALS 


| AMERICAN CONFERENCE | DRY TYPE | 
| OF GOVERNMENTAL CENTRIFUGAL COLLECTORS | 


| INDUSTRIAL HYGIENISTS bag 39a 


Industrial Ventilation 


4-24 


Q-€90G6I-d¥6 


DEVIA IONS 


a : 
hai 
1 
Bs 
J 
[:*) 


ae 


eae 


BCDF {|; 
a 


Ca ena 


Sanaa daa | S1SigiA_ NIE 


2 a Ss 


WAS e14/Snivuo NI NOLLVYINZDNOD - JONVY ONiLDSTION IsNd 


= 


— 


W42S eld ee NI tama JONVY HALT wi 


MEAN PARTICLE SIZE IN MICRONS 


RANGE OF PARTICLE SIZES, CONCENTRATION, & COLLECTOR PERFORMANCE 


RANGE OF PARTICLE SIZE 


COPYRIGHT 1952 AMERICAN AIR FILTER CO. INC. 


ACKNOWLEDGEMENTS OF PARTIAL SOURCES OF DATA REPORTED: 
2 FIRST ANO DRINKER - ARCHIVES OF INDUSTRIAL HYGIENE AND OCCUPATIONAL MEDICINE - APRIL 1952 


3 TAFT, INSTITUTE AND AAF LABORATORY TEST DATA- 1961 - “63 


4 REVERSE COLLAPSE CLOTH CLEANING ADDED 196 4 


AMERICAN CONFERENCE 


| FRANK W.G. - AMERICAN AIR FILTER - SIZE AND CHARACTERISTICS OF AIR BORNE SOLIDS - 1931 


COMPILED BY S. SYLVAN APRIL 1952 : 


p OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


Standard wet collectors are used to collect many types of 
mists. Specially designed electrostatic precipitators are fre- 
quently employed to collect sulfuric acid or oil mist. Even 
fabric and centrifugal collectors, although not the types pre- 
viously mentioned, are widely used to collect oil mist gener- 
ated by high speed machining. 


4.6 GASEOUS CONTAMINANT COLLECTORS 


Equipment designed specifically to control gas or vapor 
contaminants can be classified as: 


1. Adsorbers 
2. Thermal oxidizers 
3. Direct combustors 


4. Catalytic oxidizers 


4.6.1 Absorbers: Absorbers remove soluble or chemi- 
cally reactive gases from an air stream by contact with a 
suitable liquid. While all designs utilize intimate contact 
between the gaseous contaminant and the absorbent, different 
brands vary widely in configuration and performance. Re- 
moval may be by absorption if the gas solubility and vapor 
pressure promote absorption or chemical reaction. Water is 
the most frequently used absorbent, but additives are fre- 
quently required. Occasionally other chemical solutions must 
be used. Packed towers (Figure 4-14) are typical absorbers. 


4.6.2 Adsorbers: Adsorbers remove contaminants by col- 
lection on a solid. No chemical reaction is involved as adsorp- 
tion is a physical process where molecules of a gas adhere to 
surfaces of the solid adsorbent. Activated carbon or molecular 
sieves are popular adsorbents. 


4.6.3 Thermal Oxidizers: Thermal oxidizers, or after- 
burners, may be used where the contaminant is combustible. 
The contaminated air stream is introduced to an open flame 
or heating device followed by a residence chamber where 
combustibles are oxidized producing carbon dioxide and 
water vapor. Most combustible contaminants can be oxidized 
at temperatures between 1000 and 1500 F. The residence 
chamber must provide sufficient dwell time and turbulence to 
allow complete oxidation. 


4.6.4 Direct Combustors: Direct combustors differ from 
thermal oxidizers by introducing the contaminated gases and 
auxiliary air directly into the burner as fuel. Auxiliary fuel, 
usually natural gas or oil, is generally required for ignition 
and may or may not be required to sustain burning. 


4.6.5 Catalytic Oxidizers: Catalytic oxidizers may be 
used where the contaminant is combustible. The contami- 
nated gas stream is preheated and then passed through a 
catalyst bed which promotes oxidation of the combustibles to 
carbon dioxide and water vapor. Metals of the platinum 
family are commonly used catalysts which will promote 


Air Cleaning Devices 4-25 


oxidation at temperatures between 700 and 900 F. 


To use either thermal or catalytic oxidation, the combusti- 
ble contaminant concentration must be below the lower ex- 
plosive limit. Equipment specifically designed for control of 
gaseous or vapor contaminants should be applied with caution 
when the air stream also contains solid particles. Solid par- 
ticulates can plug absorbers, adsorbers, and catalysts and, if 
noncombustible, will not be converted in thermal oxidizers 
and direct combustors. 


Air streams containing both solid particles and gaseous con- 
taminants may require appropriate control devices in series. 


4.7 UNIT COLLECTORS 


Unit collector is a term usually applied to small fabric 
collectors having capacities in the 200-2000 cfm range. They 
have integral air movers, feature small space requirements and 
simplicity of installation. In most applications cleaned air is 
recirculated, although discharge ducts may be used if the 
added resistance is within the capability of the air mover. One 
of the primary advantages of unit collectors is a reduction in 
the amount of duct required, as opposed to central systems, 
and the addition of discharge ducts to unit collectors negates 
that advantage. 


When cleaned air is to be recirculated, a number of precau- 
tions are required (see Chapter 7). 


Unit collectors are used extensively to fill the need for dust 
collection from isolated, portable, intermittently used or fre- 
quently relocated dust producing operations. Typically, a 
single collector serves a single dust source with the energy 
saving advantage that the collector must operate only when 
that particular dust producing machine is in operation. 


Figure 4-15 shows a typical unit collector. Usually they are 
the intermittent-duty, shaker-type in envelope configuration. 
Woven fabric is nearly always used. Automatic fabric clean- 
ing is preferred. Manual methods without careful scheduling 
and supervision are unreliable . 


4.8 DUST COLLECTING EQUIPMENT COST 


The variations in equipment cost, especially on an installed 
basis, are difficult to estimate. Comparisons can be mislead- 
ing if these factors are not carefully evaluated. 


4.8.1 Price Versus Capacity: All dust collector prices per 
cfm of gas will vary with the gas flow rate. The smaller the 
flow rate, the higher the cost per cfm. The break point, where 
price per cfm cleaned tends to level off, will vary with the 
design. See the typical curves shown on Figure 4-16. 


4.8.2 Accessories Included: Careful analysis of compo- 
nents of equipment included is very important. Some collector 
designs include exhaust fan, motor, drive, and starter. In other 
designs, these items and their supporting structure must be 
obtained by the purchaser from other sources. Likewise, while 


4-26 Industrial Ventilation 


Table 4—3. Dust Collector Selection Guide 


Concen- Particle 
Operation tration Sizes 
Note 1 Note w 
CERAMICS 
a. Raw product handling light fine 
b. Fettling light fine-. 
medium 
c. Refractory sizing heavy coarse 


d. Glaze & vitr. enamel spray moderate medium 
CHEMICALS 

a. Material handling light- fine- 
moderate medium 


b. Crushing, grinding moderate- _fine- 


heavy coarse 
c. Pneumatic conveying very fine- 
heavy coarse 
d. Roasters, kilns, coolers heavy mid- 
coarse 
COAL, MINING AND POWER PLANT 
a. Material handling moderate medium 
b. Bunker ventilation light fine 
c. Dedusting, air cleaning heavy medium- 
coarse 
d. Drying moderate fine 
FLY ASH 
a. Coal burning--chain grate light fine 
b. Coal burning--stoker fired moderate fine- 
coarse 
c. Coal burning--pulverized moderate _ fine 
d. Wood burning varies coarse 
FOUNDRY 
a, Shakeout light- fine 
moderate 
b. Sand handling moderate _fine- 
medium 
c. Tumbling mills heavy medium- 
coarse 
d. Abrasive cleaning moderate- _fine- 
heavy medium 
GRAIN ELEVATOR, FLOUR AND FEED MILLS 
a. Grain handling light medium 
b. Grain dryers light coarse 
c. Flour dust moderate § medium 
d. Feed mill moderate = medium 
METAL MELTING 
a. Steel blasi furnace heavy varied 
b. Steel open hearth moderate _fine- 
coarse 
c. Steel electric furnace light fine 
d. Ferrous cupola moderate varied 
e. Non-ferrous reverberatory varied fine 
f. Non-ferrous crucible light fine 
METAL MINING AND ROCK PRODUCTS 
a. Material handling moderate _fine- 
medium 
b. Dryers, kilns moderate medium- 
coarse 
c. Rock dryer moderate _fine- 
medium 
d. Cement kiln heavy fine- 


medium 


Collector Types Used in Industry 


Dry Cen- 
trifugal 
Collector 


ee ey ee OOnO 2 2 2 a nn AM Zz NANO Oo Oo 0 NM Zz nam 


z= ZO 2 


Wet 
Collector 


On nM O On MO 


anon OO ANnNNNM aA nan O OO nn nN Oo ONnNN 


Zz nO O 


Fabric 
Collector 


OO090 Oo OO 0 0 0 OO 0 ACN oO Oo 0 O09 O20 00 


COO00 NM 


O- a © © 


Low-Volt 
Electro- 
static 


2aza2z Zz 2222 we ae Zz Pc z= 222 za 2 Zz =z a a ae 


2 2 2 2 


Hi-Volt 
Electro- 
static 


Bie res Sa Zzaz aa 


Zz oie ee 


nan oO O00 


=a 2 2 2 


mn Nn O 2 


See 
Remark No. 


ww 


> 
N DO TO FE 


= _ 
oP ~E 


Air Cleaning Devices 4-27 


Collector Types Used in Industry 


Concen- Particle Dry Cen- Low-Volt —_Hi-Volt 
Operation tration Sizes trifugal Wet Fabric Electro- Electro- See 
Note 1 Note w Collector __ Collector _ Collector __ static static Remark No. 
e. Cement grinding moderate _ fine N N O N N 33 
f. Cement clinker cooler moderate coarse 0 N O N N 34 
METAL WORKING 49 
a. Production grinding, light coarse 0 ie) O N N 35 
scratch brushing, abrasive 
cut off 
b. Portable and swing frame light medium $ e) O N N 
c. Buffing light varied S O O N N 36 
d. Tool room light fine S S S N N 37 
e. Cast iron machining moderate varied 0) (0) O S N 38 
PHARMACEUTICAL AND FOOD PRODUCTS 
a. Mixers, grinders, weighing, _light medium 10) 10) ) N N 39 
blending, bagging, 
packaging 
b. Coating pans varied fine- N 0) O N N 40 
medium 
PLASTICS 49 
a. Raw material processing See comments under 10) S 6) N N 4 
hemicals) 
b. Plastic finishing light- varied S) $ N 42 
moderate 
c. Extrusion light fine N S N O N 
RUBBER PRODUCTS 49 
a. Mixers moderate —_ fine $ O S N N 43 
b. Batchout rolls light fine iS) 0 $ S N 
c. Talc dusting and dedusting © moderate = medium iS) S 6) N N 44 
d. Grinding moderate coarse ie) 0 O N N 45 
WOODWORKING 49 
a. Woodworking machines moderate varied 0 S 8) N N 46 
b. Sanding moderate fine S S O N N 47 
c. Waste conveying, hogs heav varied 0) S) $ N N 48 


Note 1: Light: less than 2 gr/ft®, Moderage: 2 to 5 gr/ft®; Heavy: 5 gr/ft® and up. 
Note 2: Fine: 50% less than 5 microns; Medium: 50% 5 to 15 microns; Coarse: 50% 15 microns and larger. 
Note 3: O = often; S = seldom; N = never. 


Remarks Referred to in Table 4-3 


. Dust released from bin filling, conveying, weighing, mixing, pressing 


forming. Refractory products, dry pan and screen operations more 
severe. 

Operations found in vitreous enameling, wall and floor tile, pottery. 
Grinding wheel or abrasive cut-off operation. Dust abrasive. 
Operations include conveying, elevating, mixing, screening, weighing, 
packaging. Category covers so many different materials that recom- 
mendation will vary widely. 

Cyclone and high efficiency centrifugals often act as primary collectors 
followed by fabric or wet type. 

Cyclones used as product collector followed by fabric arrester for high 
over-all collection efficiency. 

Dust concentration determines need for dry centrifugal; plant location, 
product value determines need for final collectors, High temperatures 
are usual and corrosive gases not unusual. 

Conveying, screening, crushing, unloading. 

Remove from other dust producing points. Separate collector usually. 
Heavy loading suggests final high efficiency collector for all except 
very remote locations. 


11. 


12. 


13. 


14. 
15. 
16. 
17. 
18. 


19. 
20. 


21. 


Difficult problem but collectors will be used more frequently with air 
pollution emphasis. 

Public nuisance from boiler blow-down indicates collectors are 
needed. 

Large installations in residential areas require electrostatic in addition 
to dry centrifugal. 

Cyclones used as spark arresters in front of fabric collectors. 

Hot gases and steam usually involved. 

Steam from hot sand, adhesive clay bond involved. 

Concentration very heavy at start of cycle. 

Heaviest load from airless blasting due to higher cleaning speed. 
Abrasive shattering greater with sand than with grit or shot. Amounts 
removed greater with sand castings, less with forging scale removal, 
least when welding scale is removed. 

Operations such as car unloading, conveying, weighing, storing. 
Collection equipment expensive but public nuisance complaints be- 
coming more frequent. 

Operations include conveyors, cleaning rolls, sifters, purifiers, bins 
and packaging. 


4-28 Industrial Ventilation 


Remarks Referred to in Table 4-3 (continued) 


22, Operations include conveyors, bins, hammer mills, mixers, feeders 
and baggers. 

23. Primary dry trap and wet scrubbing usual. Electrostatic is added where 
maximum cleaning required. 

24. Use of this technique declining. 

25. Air pollution standards will probably require increased usage of fabric 
arresters. 

26. CAUTION! Recent design improvements such as coke-less, plasma- 
fired type, have altered emission characteristics. 

27. Zinc oxide loading heavy during zinc additions. Stack temperatures 
high. 

28. Zinc oxide plume can be troublesome in certain plant locations. 

29. Crushing, screening, conveying involved. Wet ores often introduce 
water vapor in exhaust air. 

30. Dry centrifugals used as primary collectors, followed by final cleaner. 

31. Industry is aggressively seeking commercial uses for fines. 

32. Collectors usually permit salvage of material and also reduce nuisance 
from settled dust in plant area. 

33. Salvage value of collected material high. Same equipment used on 
raw grinding before calcining. 

34, Coarse abrasive particles readily removed in primary collector types. 

35. Roof discoloration, deposition on autos can occur with cyclones and 
less frequently with high efficiency dry centrifugal. Heavy duty air filters 
sometimes used as final cleaners. 

36. Linty particles and sticky buffing compounds can cause pluggage and 
fire hazard in dry collectors. 


dust storage hoppers are integral parts of some dust collector 
designs, they are not provided in other types. Cuct connec- 
tions between elements may be included or omitted. Recircu- 
lating water pumps and/or settling tanks may be required but 
not included in the equipment price. 


4.8.3 Installation Cost: The cost of installation can equal 
or exceed the cost of the collector. Actual cost will depend on 
the method of shipment (completely assembled, sub-assem- 
bled or completely knocked down), the location (which may 
require expensive rigging), and the need for expensive sup- 
porting steel and access platforms. Factory installed media 
will reduce installation cost. The cost can also be measurably 
influenced by the need for water and drain connections, 
special or extensive electrical work, and expensive material 
handling equipment for collection material disposal. Items in 
the latter group will often also be variable, decreasing in cost 
per cfm as the flow rate of gas to be cleaned increases. 


4.8.4 Special Construction: Prices shown in any tabula- 
tion must necessarily assume standard or basic construction. 
The increase in cost for corrosion resisting material, special 
high-temperature fabrics, insulation, and/or weather protec- 
tion for outdoor installations can introduce a multiplier of one 
to four times the standard cost. 


A general idea of relative dust collector cost is provided in 
Figure 4-16. The additional notes and explanations included 


37. Unit collectors extensively used, especially for isolated machine tools. 

38. Dust ranges from chips to fine floats including graphitic carbon. Low voltage 
ESP applicable only when a coolant is used. 

39. Materials vary widely. Collector selection depends on salvage value, 
toxicity, sanitation yardsticks. 

40, Controlled temperature and humidity of supply air to coating pans 
makes recirculation desirable. 

41. Plastic manufacture allied to chemical industry and varies with opera- 
tions involved. 

42. Operations and collector selection similar to woodworking. See Item 
13. 

43. Concentration is heavy during feed operation. Carbon black and other 
fine additions make collection and dust-free disposal difficult. 

44. Salvage of collected material often dictates type of high efficiency 
collector. 


45. Fire hazard from some operations must be considered. 

46. Bulking material. Collected material storage and bridging from splin- 
ters and chips can be a problem. 

47. Dry centrifugals not effective on heavy concentration of fine particles 
from production sanding. 

48. Dry centrifugal collectors required. Wet or fabric collectors may be 
used for final collectors. 


49. See NFPA publications for fire hazards, e.g., zirconium, magnesium, 
aluminum, woodworking, plastics, etc. 


in these data should be carefully examined before they are 
used for estimating the cost of specific installations. For more 
accurate data, the equipment manufacturer or installer should 
be asked to provide estimates or a past history record for 
similar control problems utilized. Table 4-4 lists other char- 
acteristics that must be evaluated along with equipment cost. 


Price estimates included in Figure 4-16 are for equipment 
of standard construction in normal arrangement. Estimates for 
exhausters and dust storage hoppers have been included, as 
indicated in Notes 1 and 2, where they are normally furnished 
by others. 


4.9 SELECTION OF AIR FILTRATION EQUIPMENT 


Air filtration equipment is available in a wide variety of 
designs and capability. Performance ranges from a simple 
throwaway filter for the home furnace to the "clean room" in 
the electronics industry, where the air must be a thousand 
times as clean as in a hospital surgical suite. Selection is based 
on efficiency, dust holding capacity, and pressure drop. There 
are five basic methods of air filtration. 


4.9.1 Straining: Straining occurs when a particle is larger 
than the opening between fibers and cannot pass through. It 
is a very ineffective method of filtration because the vast 
majority of particles are far smaller than the spaces between 
fibers. Straining will remove lint, hair, and other large parti- 


Air Cleaning Devices 4-29 


—- Acoustical 
enclosure 


}-—— Fan Impeller 


Scroll 


Automatic — 
shaker 
controls ; 
Shaker JF 


motor od 


Envelopes 


e-————-  Air inlet 


Shaker —- 


Funnel hopper 


Dust drum 


UNIT COLLECTOR 
(FABRIC-SHAKER TYPE) 


| AMERICAN CONFERENCE 
# OF GOVERNMENTAL | | 
| INDUSTRIAL HYGIENISTS 


4-30 Industrial Ventilation 


RELATIVE COST, PERCENT 


10 
CFM IN THOUSANDS 


High voltage precipitator (minimum cost range) 
Continuous duty high temperature fabric collector (2.0:1) 
Continuous duty reverse pulse (8:1) 

Wet collector 

Intermittent duty fabric collector (2.0:1) 

Low voltage precipitator 

Cyclone 


Cost based on collector section only. Does not include ducts, dust disposal 
devices, pumps, exhausters or other accessories not an integral part of the 
collector. 


Price of high voltage precipitator will vary substantially with applications 
and efficiency requirements. Costs shown are for fly ash aplications 
where velocities of 200 to 300 fpm are normal. 


AMERICAN CONFERENCE | COST ESTIMATES OF 
OF GOVERNMENTAL §DUST COLLECTING EQUIPMENT 


TABLE 4-4. Comparison of Some Important Dust Collector Characteristics 


Type 


Electrostatic 
Fabric 
Intermittent—Shaker 
Continuous—Shaker 
Continuous—Reverse Air 
Continuous—Reverse Pulse 
Glass, Reverse flow 
Wet: 
Packed Tower 
Wet Centrifugal 
Wet Dynamic 
Orifice Types 


Higher Efficiency: 
Fog Tower 
Venturi 

Dry Centrifugal: 
Low Pressure Cyclone 
High Eff. Centrifugal 
Dry Dynamic 


Higher efficiency 
Range on Particles 


Greater than Mean Pressure H,0 Gal per 
Size in Microns Loss Inches 1,000 cfm Space 
0.25 Yo _— Large 
0.25 3-6 — Large 
0.25 3-6 —_ Large 
0.25 gen Ne o Large 
0.25 3-6 — Moderate 
0.25 3-8 — Large 
1-5 1.5-3.5 5-10 Large 
1-5 2.5-6 3-5 Moderate 
1-2 Note 2 ‘e to 1 Small 
1-5 242-6 10-40 Small 
0.5-5 2-4 5-10 Moderate 
0.5-2 10-100 5-15 Moderate 
20-40 0.75-1.5 —_ Large 
10-30 3-6 _— Moderate 
10-20 Note 2 _— Small 


Sensitivity to cfm Change 


Pressure Efficiency 
Negligible Yes 
As cfm Negligible 
As cfm Negligible 
As cfm Negligible 
As cfm Negligible 
As cfm Negligible 
As cfm Yes 
As (cfm)? Yes 
Note 2 No 
As cfm or less Varies with 
design 
As (cfm)? Slightly 
As (cfm)" Yes 
As (cfm)? Yes 
As (cfm)? Yes 
Note 2 No 


Note 1: Pressure loss is that for fabric and dust cake. Pressure losses associated with outlet connections to be added by system designer. 


Note 2: A function of the mechanical efficiency of these combined exhausters and dust collectors. 


Note 3: Precooling of high temperature gases will be necessary to prevent rapid evaporation of fine droplets. 
Note 4: See NFPA requirements for fire hazards, e.g., zirconium, magnesium, aluminum, woodworking, etc. 


Humid Air Influence 


Improves Efficiency 
May Make 


Reconditioning 
Difficult 


None 


None 


May Cause 
condensation 
and plugging 


Max. Temp. F 
Standard 
Construction 
Note 4 


500 


See Table 4-1 


550 


Unlimited 


Note 3 
Unlimited 


750 
750 


sadiAag SulUBIID ALY 


le-v 


4-32 Industrial Ventilation 


cles. 


4.9.2 Impingement: When air flows through a filter, it 
changes direction as it passes around each fiber. Larger dust 
particles, however, cannot follow the abrupt changes in direc- 
tion because of their inertia. As a result, they do not follow 
the air stream and collide with a fiber. Filters using this 
method are often coated with an adhesive to help fibers retain 
the dust particles that impinge on them. 


4.9.3 Interception: Interception is a special case of im- 
pingement where a particle is small enough to move with the 
air stream but, because its size is very small in relation to the 
fiber, makes contact with a fiber while following the tortuous 
air flow path of the filter. The contact is not dependent on 
inertia and the particle is retained on the fiber because of the 
inherent adhesive forces that exist between the particle and 
fiber. These forces, called van der Waals (J. D. van der Waals, 
1837-1923) forces, enable a fiber to trap a particle without 
the use of inertia. 


4.9.4 Diffusion: Diffusion takes place on particles so 
small that their direction and velocity are influenced by 
molecular collisions. These particles do not follow the air 
stream, but behave more like gases than particulate. They 
move across the direction of air flow in a random fashion. 
When a particle does strike a fiber, it is retained by the van 
der Waals forces existing between the particle and the fiber. 
Diffusion is the primary mechanism used by most extremely 
efficient filters. 


4.9.5 Electrostatic: A charged dust particle will be at- 
tracted to a surface of opposite electrical polarity. Most dust 
particles are not electrically neutral; therefore, electrostatic 
attraction between dust particle and filter fiber aids the col- 
lection efficiency of all barrier-type air filters. Electrostatic 
filters establish an ionization field to charge dust particles so 
that they can be collected on a surface that is grounded or of 
opposite polarity. This concept was previously discussed in 
Section 4.3.1. 


Table 4-5 shows performance versus filter fiber size for 
several filters. Note that efficiency increases as fiber diameter 
decreases because more small fibers are used per unit volume. 


ARRESTANCE 
92-76 


EFFICIENCY 


52-76 


DOP 


Note also that low velocities are used for high-efficiency 
filtration by diffusion. 


The wide range in performance of air filters makes it 
necessary to use more than one method of efficiency testing. 
The industry-accepted methods in the United States are 
ASHRAE Arrestance, ASHRAE Efficiency, and DOP. For 
ASHRAE Arrestance, a measured quantity of 72% stand- 
ardized air cleaner test dust, 23% carbon black, and 5% cotton 
lint is fed to the filter. The efficiency by weight on this specific 
test dust is the ASHRAE Arrestance. ASHRAE Efficiency is 
a measure of the ability of a filter to prevent staining or 
discoloration. It is determined by light reflectance readings 
taken before and after the filter in a specified test apparatus. 
Atmospheric dust is used for the test. Both ASHRAE tests are 
described in ASHRAE Publication 52-76.4%) 


In a DOP Test, 0.3 micron particles of dioctylphthalate 
(DOP) are drawn through a HEPA (High Efficiency Particu- 
late Air) filter. Efficiency is determined by comparing the 
downstream and upstream particle counts. To be designated 
as a HEPA filter, the filter must be at least 99.97% efficient, 
i.e., only three particles of 0.3 micron size can pass for every 
ten thousand particles fed to the filter. Unlike both ASHRAE 
tests, the DOP test is not destructive, so it is possible to repair 
leaks and retest a filter that has failed. 


The three tests are not directly comparable; however, Fig- 
ure 4-17 shows the general relationship. Table 4-6 compares 
several important characteristics of commonly used air filters. 
Considerable life extension of an expensive final filter can be 
obtained by the use of one or more cheaper, less efficient, 
prefilters. For example, the life of a HEPA filter can be 
increased 25% with a throwaway prefilter. If the throwaway 
filter is followed by a 90% efficient extended surface filter, 
the life of the HEPA filter can be extended nearly 900%. This 
concept of "progressive filtration" allows the final filters in 
clean rooms to remain in place for ten years or more. 


The European Committee on the Construction of Air Han- 
dling Equipment has developed a method for testing air filters 
in general ventilation. Although their method, called 
Eurovent 4/5, is based directly on ASHRAE Standard 52-76, 
some wording and definitions have been amended to suit the 
needs of Eurovent. Eurovent 4/5 aims to establish a uniform 


FIGURE 4-17. Comparison between various methods of measuring air cleaning capability. 


comparative testing procedure for air filters having volumet- 
ric flow rates greater than 0.236 m?/s (500 cfm) and an average 
dust spot efficiency up to 98%. 


The wide range of filter efficiency is segregated into 14 
grades of filters from EU! to EU14. 


4.10 RADIOACTIVE AND HIGH TOXICITY OPERATIONS 


There are three major requirements for air cleaning equipment 
to be utilized for radioactive or high toxicity applications: 


1. High efficiency 
2. Low maintenance 
3. Safe disposal 


High efficiency is essential because of extremely low tol- 
erances for the quantity and concentration of stack effluent 
and the high cost of the materials handled. Not only must the 
efficiency be high, it must also be verifiable because of the 
Jegal requirement to account for all radioactive material. 


The need for low maintenance is of special importance 
when exhausting any hazardous material. For many radioac- 
tive processes, the changing of bags in a conventional fabric 
collector may expend the daily radiation tolerances of 20 or 
more persons. Infrequent, simple, and rapid maintenance 
requirements are vital. Another important factor is the desir- 
ability of low residual buildup of material in the collector 
since dose rates increase with the amount of material and 
reduce the allowable working time. 


Disposal of radioactive or toxic materials is a serious and 
very difficult problem. For example, scalping filters loaded 
with radioactive dust are usually incinerated to reduce the 
quantity of material that must be disposed of in special burial 
grounds. The incinerator will require an air cleaning device, 
such as a wet collector of very special design, to avoid 
unacceptable pollution of air and water. 


With these factors involved, it is necessary to select an air 
cleaning device that will meet efficiency requirements with- 
out causing too much difficulty in handling and disposal. 


Filter units especially designed for high efficiency and low 
maintenance are available. These units feature quick 
changeout through a plastic barrier which is intended to 
encapsulate spent filters, thereby eliminating the exposure of 
personnel to radioactive or toxic material. A filtration effi- 
ciency of 99.97% by particle count on 0.3 micron particles is 
standard for this type of unit. 


For further information on this subject, see Reference 4.5. 
4.11 EXPLOSION VENTING 


There is a wide range of dusts which are combustible and 
capable of producing an explosion. Explosions occur when 
the right concentration of finely divided dust is suspended in 
air and exposed to a sufficient source of ignition. A dust 
collector, by its very operation, maintains a cloud of finely 


Air Cleaning Devices 4-33 


TABLE 4-5. Media Velocity vs. Fiber Size 


Media 
Filter Size Velocity _ Filtration 
Filter Type (microns) (fpm) | Mechanism 
Panel Filters 25-50 250-625 Impingement 
Automatic Roll Filters 25-50 500 Impingement 
Extended Surface Filters 0.75-2.5 20-25 Interception 
HEPA Filters 0.5-6.3 5 Diffusion 


divided particles suspended in air. If a source of ignition 
initiates the combustion of the dust cloud, the gases in the 
cloud will rapidly expand due to heat developed during the 
combustion. If a dust collector vessel constructs this expan- 
sion, a rapid pressure buildup inside the collector casing will 
cause a violent rupture. When dust particles are know to be 
combustible, precautions for an explosion must be taken and 
suitable protection provided to reduce the risk of property 
damage and personal injury. 


To begin taking precautions, sources of possible ignition 
must be identified and controlled to minimize the risk of a 
dust cloud explosion. Usual causes of explosions include 
static by minimizing the ignition sources such as static dis- 
charge, hot surfaces on machinery and sparks and flames from 
processes. After identifying possible sources of ignition, pre- 
ventive measures should be taken. Static grounding of the 
equipment and spark traps are typical preventive measures. 
The addition of an inert gas to replace oxygen in a dust 
collector can prevent an explosion by ensuring the minimum 
oxygen content required for ignition is never reached. Inerting 
can be very effective in closed loop systems but is not eco- 
nomical in typical local exhaust systems because of the con- 
stant loss of expensive inerting gas. Should ignition occur, 
protective measures must be taken to limit the damage. Typi- 
cal protective measures include explosion suppression, explo- 
sion containment, and explosion venting. 


Explosion suppression requires the early detection of an 
explosion, usually within the first 20 milliseconds. Once 
ignition is detected, an explosion suppression device injects 
a pressurized chemical suppressant into the collector to dis- 
place the oxygen and impede combustion. These are typically 
used in conjunction with fast acting isolation valves on the 
inlet and outlet ducts. These systems can be very useful when 
toxic dusts are being handled. 


Explosion containment uses specialized dust collectors 
designed to withstand the maximum pressure generated and 
contain the explosion. Most pressure capabilities of commer- 
cially available dust collectors are not sufficient to contain an 
explosion in progress. 


Explosion venting, the most common protection, is af- 
forded by fitting pressure relief vents to the collector housing. 


4-34 Industrial Ventilation 


TABLE 4-6. Comparison of Some Important Air Filter Characteristics* 


Pressure Drops "wg ASHRAE Performance Maintenance 
(Notes 1 & 2) (Note 4) (Note 5) 
Face Velocity 
Type Initial Final Arrestance Efficiency fpm Labor Material 
LowMedium Efficiency 
1. Glass Throwaway 0.1 0.5 77% NA 300 High High 
(2' deep) Note 6 
2. High Velocity 0.1 0.5 73% NA 500 High Low 
(permanent units) Note 6 
(2" deep) 
3. Automatic 0.4 0.4 80% NA 500 Low Low 
(viscous) Note 6 
Medium/High Efficiency 
1. Extended Surface 0.15-0.60 0.5-1.25 90-99% 25-95% 300-625 Medium Medium 
(dry) 
2. Electrostatic: 
a. Dry Agglomerator/ 0.35 0.35 NA 90% 500 Medium Low 
Roll Media Note 7 
b. Dry Agglomerator/ 0.55 1.25 NA 95%+ 530 Medium Medium 
Extended Surface Note 7 
Media 
c. Automatic Wash 0.25 0.25 NA 85-95% 400-600 Low Low 
Type Note 7 
Ultra High Efficiency 
1. HEPA 0.5-1.0 1.0-3.0 Note 3 Note 3 250-500 High High 


Note 1: Pressure drop values shown constitute a range or average, whichever is applicable. 


Note 2: Final pressure drop indicates point at which filter or filter media is removed and the media is either cleaned or replaced. All others are 
cleaned in place, automatically, manually or media renewed automatically. Therefore, pressure drop remains approximately constant. 


Note 3: 95-99.97% by particle count, DOP test. 


Note 4: ASHRAE Standard 52-76 defines (a) Arrestance as a measure of the ability to remove injected synthetic dust, calculated as a percentage on a weight basis and 
(b) Efficiency as a measure of the ability to remove atmospheric dust determined on a light-transmission (dust spot) basis. 


Note 5: Compared to other types within efficiency category. 
Note 6: Too low to be meaningful. 
Note 7: Too high to be meaningful. 


As pressure increases quickly leading up to an explosion, a 
relief vent opens to allow the rapidly expanding gases to 
escape. This effectively limits the maximum pressure build- 
up to less than the bursting pressure of the vessel. The neces- 
sary area for such a relief vent is a function of the vessel 
volume, vessel strength, the opening pressure of the relief vent 
and the rate of pressure rise characteristic of the dust in 
question. Most standard dust collectors will require reinforc- 
ing to withstand the reduced maximum pressure experienced 
during an explosion. 


To choose the most reliable, economical and effective 
means of explosion control, an evaluation of the specifics of 
the exhaust system and the degree of protection required is 
necessary. 


NAPA 68-1994, Guide for Explosion Venting,(4.6) is the 
most commonly recognized standard and should be studied 


and thoroughly familiar to anyone responsible for the design 
or evaluation of dust collectors applied to potentially explo- 
sive dusts. 


REFERENCES 


4.1. Leith, D.; First, M.K.W.; Feldman, H.: Performance 
of a Pulse-Jet at High Velocity Filtration I, Filter Cake 
Redeposition. J. Air Pollut. Control Assoc. 28:696 
(July 1978). 

4.2. Beake, E.: Optimizing Filtration Parameters. J. Air 
Pollut. Control Assoc. 24:1150 (1974). 


4.3. Leith, D.; Gibson, D.D.; First, M.W.: Performance of 
Top and Bottom Inlet Pulse-Jet Fabric Filters. J. Air 
Pollut. Control Assoc. 24:1150 (1974). 


4.4. American Society of Heating, Refrigerating and Air- 
Conditioning Engineers: Method of Testing Cleaning 


Air Cleaning Devices 4-35 


Devices Used in General Ventilation for Removing tection Criteria. NCRP Report No. 39. NCRP Publi- 

Particulate Matter. ASHRAE Pub. No. 52-76. cations, Bethesda, MD (January 1971). 

ASHRAE, Atlanta, GA (May 1976). 4.6. National Fire Protection Association: Guide for Ex- 
4.5. National Council on Radiation Protection and Meas- plosion Venting. NFPA 68-1978. NFPA, Quincy, MA 


urement: NCRP Report No. 39, Basic Radiation Pro- (1978). 


Chapter 5 


EXHAUST SYSTEM DESIGN PROCEDURE 


5.1 INTRODUCTION ...............-..00. 5-2 5.12. FRICTION LOSS FOR NON-CIRCULAR DUCTS. 5-13 
5.2 PRELIMINARY STEPS................-. 5-2 5.13 CORRECTIONS FOR NONSTANDARD DENSITY 5-15 
5.3 DESIGN PROCEDURE........-......... 5-2 5.13.1 Variable Temperature and/or Different 
5.4 DUCT SEGMENT CALCULATIONS ......... 5-3 Altitude 22.2. ee ee 5-27 
5.5 DISTRIBUTION OF AIRFLOW. .........-. 5-4 5.13.2 Elevated Moisture... 1.6... 0-5... 5-27 
5.5.1 Balance by Design Method. .......... 5-4 5.13.3. Psychrometric Principles 2... ....... 5-27 
55:2. Blast'Gate Method: 2.03< fu 4a sky Po 5-10 5.13.4 Density Determination... ... 2... -.- 5-28 
553  ChoiceofMethods .............. 5-10 5.13.5 Hood Flow Rate Changes with Density . . . . 5-28 
5.5.4 Balance by Design Procedure... ...... 5-10 5.14 AIR CLEANING EQUIPMENT ............ 5-32 
5.5.5 Blast Gate Procedure... 2... 2. 2... 5-10 5:15 EVASE DISCHARGE yo. suc4od 4) parent ened 5-32 
5.5.6  SystemRedesign ............... 5-10 5.16 EXHAUST STACK OUTLETS ............ 5-33 
56 AIDSTOCALCULATION.............. 5-11 5.16.1 Stack Considerations ............. 5-34 
5.7 PLENUMEXHAUST SYSTEMS .......... 5-11 5.17 AIRBLEED-INS .. 2... 2.0.2 es 5-35 
5.7.1 Choice of Systems .............. 5-11 5.18 OPTIMUM ECONOMIC VELOCITY ........ 5-35 
BID DES Of Rh tent, Bub he a oe 5-1] 5.19 CONSTRUCTION GUIDELINES FOR LOCAL 
5.8 FAN PRESSURE CALCULATIONS ........ 5-11 EXOT UR MEMS cai tw atajehintae a sie 9 BES 
5.8.1 Fan Total Pressure .............. 5-11 5.19.1 Meee fa pe es A ah Se = 
5.8.2 Fan Static Pressure 2... 5-12 sacs Construction SCY We Rae pe Oe SEN Fey aes 
5.8.3 Completion of the Example on Figure 5-3. . 5-12 DAP SS MIS aid 2h at ke tee pe eg 4 8 
5.9 CORRECTIONS FOR VELOCITY CHANGES . . . 5-12 PDD SOOUS: Stee Gas TOES act BA 2338 
£O'|- “Branch Eaiesio Man Duce: oe 5-12 5.19.5 De Types of Duct Materials ........ 5-38 
5.9.2 Contractions and Expansions ......... 5-13 5.19.6 Testing... 2. eee eee pene 
5.10 SAMPLESYSTEMDESIGN............. Seige. EOE ES rent ie ae 2 By Stat he a8 
5.11. DIFFERENT DUCT MATERIAL FRICTION 
LOSSES * sc\jan08 Bivorcn db O88 Bite eta 5-13 
Figure 5-1 System Duct Calculation Parameter Location . . 5-3 Figure 5-17 — Psychrometric Chart—Normal Temperature . . 5-54 
Figure 5-2 Problem: 1)... 2 ais wd ake, aoe wae at ee ks 5-5 Figure 5-18 = Psychrometric Chart—Low Temperatures . 5-55 
Figure 5-3 Velocity Pressure Method Calculation Sheet .. . 5-6 Figure 5-19 Psychrometric Chart for High Temperatures . . 5-56 
Figure 5-4 Plenum Vs. Conventional System ........ 5-8 Figure 5-20 Psychrometric Chart for Very High 
Figure 5-5 Types of Plenums................. 5-9 Temperatures... 2... 2... ee 5-57 
Figure 5-6 Branch Entry Velocity Correction ........ 5-13 Figure 5-21 Principles of Duct Design Elbows ....... 5-58 
Figure 5-7 Expansions and Contractions ........... 5-14 Figure 5-22 Heavy Duty Elbows... 2. .......... 5-59 
Figure 5-8 Problerin 23 05. Sss Stash wee oR Gok eet ak as 5-16 Figure 5-23. Cleanout Openings ..........0....-. 5-60 
Figure 5-9 Balanced Design Method Calculation Sheet . . . 5-17 Figure 5-24 Blast Gates: j.-c2 te Pk mec Mv ee ee 4 be 5-61 
Figure 5-10 — Blast Gate Method Calculation Sheet... 2... 5-23 Figure 5-25 Principles of Duct Design... 2.2... 2... 5-62 
Figure 5-11 System Layout .. 2... 2. .0.0...0.00.- 5-28 Figure 5-26 Principles of Duct Design Branch Entry .... 5-63 
Figure 5-12 Psychrometric Chart for Humid Air 2... 2... 5-30 Figure 5-27 Principles of Duct Design Fan Inlets . 2... . 5-64 
Figure 5-13. Hood Entry Loss Coefficients .......... 5-40 Figure 5-28 Airflow Around Buildings ........... 5-65 
Figure 5-14. Duct Design Data Elbow Losses ..... 2... 5-41 Figure 5-29 _—_ Effective Stack Height and Wake Downwash 5-66 
Figure 5-15 Duct Design Data (Branch Entry Losses/ Figure 5-30 = Stackhead Designs ............... 5-67 
Weather Cap Losses)... .......-.0.0.. 5-42 
Figure 5-16 | Duct Design Data (Static Pressure 
Regains/Losses) .......-....--2-.- 5-47 


5-2 industrial Ventilation 


5.1 INTRODUCTION 


The duct system that connects the hoods, air cleaning 
device(s), and fan must be properly designed. This process is 
much more involved than merely connecting pieces of duct. 
If the system is not carefully designed in a manner which 
inherently ensures that the design flow rates will be realized, 
contaminant control may not be achieved. 


The results of the following design procedure will deter- 
mine the duct sizes, material thickness, and the fan operating 
point (system flow rate and required pressure) required by the 
system. Chapter 6 describes how to select a fan based on these 
results. 


5.2 PRELIMINARY STEPS 


Coordinate design efforts with all personnel involved, in- 
cluding the equipment or process operator as well as mainte- 
nance, health, safety, fire, and environmental personnel. The 
designer should have, at a minimum, the following data 
available at the start of the design calculations: 


|. A layout of the operations, workroom, building (f 
necessary), etc. The available location(s) for the air 
cleaning device and fan should be determined. An 
important aspect that must be considered at this time 
is to locate the system exhaust point (where the air exits 
the system) so that the discharged air will not re-enter 
the work space, either through openings in the building 
perimeter or through replacement air unit intakes. (See 
Figures 5-28 and 5-29.) 


2. A line sketch of the duct system layout, including plan 
and elevation dimensions, fan location, air cleaning 
device location, etc. Number, letter, or otherwise iden- 
tify each branch and section of main duct on the line 
sketch for convenience. The examples show hoods 
numbered and other points lettered. 


Locate the fan close to pieces of equipment with high 
losses. This will facilitate balancing and may result in 
lower operating costs. 


Flexible duct is susceptible to sagging and excessive 
bending, which increases static pressure losses. Usually, 
these additional System Pressure (SP) losses cannot be 
predicted accurately. Use hard duct whenever possible 
and keep flexible duct lengths as short as possible. 


3. A design or sketch of the desired hood for each opera- 
tion with direction and elevation of outlet for duct 
connection. 


4. Information about the details of the operation(s), spe- 
cifically toxicity, ergonomics, physical and chemical 
characteristics, required flow rate, minimum required 
duct velocity, entry losses, and required capture ve- 
locities. 


5. Consider the method and location of the replacement 


air distribution devices on the hood's performance. The 
type and location of these fixtures can dramatically 
lower contaminant control by creating undesirable 
turbulence at the hood (see Chapter 7). Perforated 
plenums or perforated duct provide better replacement 
air distribution with fewer adverse effects on hood 
performance. 


5.3 DESIGN PROCEDURE 


All exhaust systems are comprised of hoods, duct seg- 
ments, and special fittings leading to an exhaust fan. A com- 
plex system is merely an arrangement of several simple 
exhaust systems connected to a common duct. There are two 
general classes of duct system designs: tapered systems and 
plenum systems. The duct in a tapered system gradually gets 
larger as additional flows are merged together, thus keeping 
duct velocities nearly constant. If the system transports par- 
ticulate (dust, mist, or condensable vapors), the tapered sys- 
tem maintains the minimum velocity required to prevent 
settling. The duct in a plenum system (see Section 5.7) is 
generally larger than that in a tapered system, and the velocity 
in it is usually low. Any particulate in the air stream can settle 
out in the large ducts. Figures 5-4 and 5-5 illustrate design 
alternatives. Regardless of which system is used, the follow- 
ing procedure will result in a workable system design. 


1. Select or design each exhaust hood based on the tox- 
icity, physical, and chemical characteristics of the 
material and the ergonomics of the process and deter- 
mine its design flow rate, minimum duct velocity, and 
entry losses (see Chapters 3 and 10). Note that mini- 
mum duct velocity is only important for systems trans- 
porting particulate, condensing vapors, or mist and to 
prevent explosive concentrations building up in the 
duct (see Section 5.18 for a discussion on economic 
velocities for non-particulate systems). 


2. Start with the duct segment that has the greatest 
number of duct segments between it and the fan. A duct 
segment is defined as the constant diameter round (or 
constant area rectangular) duct that separates points of 
interest such as hoods, entry points, fan inlet, etc. 


3. Determine the duct area by dividing the design flow 
rate by the minimum duct velocity. Convert the resul- 
tant cross-sectional area into a tentative duct diameter. 
A commercially available duct size (see Table 5-8) 
should be selected. If solid particulates or condensable 
vapors are being transported through the system, a 
minimum velocity is required (see Chapters 3 and 10). 
If the tentative duct diameter is not a standard size, 
select the next smaller size to ensure that the actual 
duct velocity is equal to or greater than the minimum 
required. 


4. Using the line sketch, determine the design length for 
each duct segment and the number and type of fittings 


(elbows, entries, and other special fittings) needed. 
Design length is the centerline distance along the duct 
(the distance between the intersection of the center- 
lines of the straight duct components). 


5. Calculate the pressure losses for the duct segments that 
merge at a common junction point. (See Section 5.4 
for the details on how to calculate these losses.) 


6. Directly after each junction point, there must be one 
and only one SP, regardless of the path taken to reach 
that point. If not ensured by the design process, the 
system will "self-balance” by reducing the flow rate in 
the higher-resistance duct segment(s) and increasing 
the flow rate in the lower-resistance duct segment(s) 
until there is a single SP in the duct downstream of 
each junction point. 


SP balance at any junction point can be achieved by 
either one of two fundamental design methods: 1) 
Adjust the flow rate through the hood(s) until the SPs 
at each junction point are the same. 2) Increase the 
resistance in the low resistance duct segment(s) by 
means of some artificial device such as a blast gate, 
orifice plate, or other obstruction in the segment. 


Section 5.5 discusses the details of these procedures. 


7. Select both the air cleaning device and fan based upon 
final calculated system flow rate, temperature, moisture 


Pee 


Exhaust System Design Procedure 5-3 


condition, contaminant loading, physical and chemical 
characteristics, and overall system resistance. 


8. Check the duct sizes designed against the available 
space and resolve any interference problems. (For 
example, will the elbow size desired actually fit in the 
available space?) This may cause a redesign of part of 
the system. 


9. Determine the material type and thickness (gauge) for 
each duct segment based on the air stream characteristics. 


5.4 DUCT SEGMENT CALCULATIONS 


The Velocity Pressure (VP) Method is based on the fact 
that all frictional and dynamic (fitting) losses in ducts and 
hoods are functions of the velocity pressure and can be 
calculated by a loss coefficient multiplied by the velocity 
pressure. Loss coefficients for hoods, straight ducts, elbows, 
branch entries, contractions, and expansions are shown in 
Figures 5-13 through 5-16. Figure 5-1 shows the application 
of these coefficients. For convenience, loss coefficients for 
round elbows and entries are also presented on the calculation 
sheet (see Figure 5-3). 


Friction data for this method are presented as Tables 5-5 
and 5-6. These tables give the loss coefficients per foot of 
galvanized and commercial steel, aluminum, PVC, and stain- 
less steel duct. The equations for these tables are listed on 


as (2) 
SP, = SP -(VP, - VP) 
Sh 
V, V3 
VP VP, 


VR 


(1) See 3.5.1 and 3.5.2 
(2) See 5.9.1 


FIGURE 5-1. System duct calculation parameter location 


duct 2 


5-4 Industrial Ventilation 


these tables and also on the calculation sheet (see Figure 
5-3). These equations and the resultant tables have been 
designed to be no more than 4% different from the "exact" 
values of the Colebrook—White equation and were designed 
to err on the high side of the normal velocity range of exhaust 
ventilation systems. 


For convenience, two data sets determined from the same 
equations were used to generate the friction tables. These 
tables are possible because, for a specific diameter, the friction 
loss coefficient changes only slightly with velocity. Each table 
lists the friction coefficient as a function of diameter for six 
different velocities. The error in using these data with veloci- 
ties plus or minus 1000 fpm is within 6%. If desired, a linear 
interpolation between velocity values can be performed. 


In Chapter 1, an equation was presented for flexible duct 
with the wires covered. No data are presented here for this 
type of material due to the wide variability from manufacturer 
to manufacturer. Perhaps an even more important reason is 
that these data are for straight duct losses, and flexible duct, 
by its very nature, is seldom straight. Typically, bends in 
flexible duct can produce extremely large losses which cannot 
be predicted easily. Be very careful to keep the flexible duct 
as straight and as short as possible. 


The following steps will establish the overall pressure loss 
of a duct segment that starts at a hood. Figure 5-2 shows a 
simple one-hood ventilation system. The use of a calculation 
sheet can be very beneficial when performing the calculations 
manually. Figure 5-3 shows the details of the calculations for 
each component of the system. There is also a profile through 
the system showing the magnitude and relationships of total, 
static, and velocity pressures on both the "suction" and the 
"oressure" sides of the fan on Figure 5-2. It should be noted 
that VP is always positive. Also, while total and static pressure 
may be either negative or positive with respect to atmospheric 
pressure, Total Pressure (TP) is always greater than SP (TP = 
SP + VP). 


NOTE: The numbers in the problems presented in this 
chapter were generated using one of the available 
computer programs (see Section 5.6). The values pre- 
sented in the calculation sheets may be different from 
those determined by other methods. 


1. Determine the actual velocity by dividing the flow rate 
by the area of the commercial duct size chosen. Then 
determine the corresponding velocity pressure from 
Table 5-7 or the equations in Chapter |. In the example, 
the diameter chosen was 4" (line 5), the actual velocity 
is given on line 7 and the VP corresponding to this 
actual velocity is given on line 8. 


2. Determine the hood static pressure from the equations 
in Chapter 3. In this example, there are no slots, so the 
duct entry loss is as given on lines 17 through 22. 


3. Multiply the design duct length by the loss coefficient 
from the tabulated data of Tables 5-5 or 5-6 (lines 23 
through 25.) The use of galvanized sheet metal duct 
was assumed throughout this chapter. 


4. Determine the number and type of fittings in the duct 
segment. For each fitting type (see Figures 5-13, 5-14, 
5-15, and 5-16), determine the loss coefficient and 
multiply by the number of fittings (there were none in 
this example.) 


5. Add the results of Steps 3 and 4 above and multiply by 
the duct VP. This is the actual loss in inches of water 
for the duct segment (given on line 34). 


6. Add the result of Step 5 to the hood suction. If there 
are any additional losses (expressed in inches of 
water), such as for an air cleaning device, add them in 
also. This establishes the cumulative energy required, 
expressed as static pressure, to move the design flow 
rate through the duct segment (line 37). Note that the 
value on line 37 is negative. 


The calculations listed in the last three columns of Figure 
5-3 will be discussed in Section 5.8.3. 


5.5 DISTRIBUTION OF AIR FLOW 


As discussed previously, a complex exhaust system is 
actually a group of simple exhaust systems connected to a 
common main duct. Therefore, when designing a system of 
multiple hoods and branches, the same rules apply. In a 
multiple branch system, however, it is necessary to provide a 
means of distributing air flow between the branches either by 
balanced design or by the use of blast gates. 


Air will always take the path of least resistance. A natural 
balance at each junction will occur; that is, the exhaust flow 
rate will distribute itself automatically according to the pres- 
sure losses of the available flow paths. The designer must 
provide distribution such that the design air flow at each hood 
will never fall below the minimums listed in Chapter 3 and/or 
Chapter 10. To do so, the designer must make sure that all 
flow paths (ducts) entering a junction will have equal calcu- 
lated static pressure requirements. 


To accomplish this, the designer has a choice of two 
methods. The object of both methods is the same: to obtain 
the desired flow rate at each hood in the system while main- 
taining the desired velocity in each branch and main. 


The two methods, labeled Balance by Design Method and 
Blast Gate Method, are outlined below. Their relative advan- 
tages and disadvantages can be found in Table 5-1. 


5.5.1 Balance by Design Method: This procedure (see 
Section 5.10) provides for achievement of desired air flow (a 
"balanced" system) without the use of blast gates. It is often 
called the "Static Pressure Balance Method.” In this type of 
design, the calculation usually begins at the hood farthest from 


Exhaust System Design Procedure 


Fabric Vertical discharge cap ~ 
ollector : 


(Fig. 5 30,5~31) 


Hood 


suction 


Details of Ope ration 


VS~ 


PRINT : 


HOOD 
NO, 


6° Diameier Grinding 
wheel, 2° Wide 


Dimensions 


or Main 


CFM 
Requirec 


15 
Collector 

] 

10 
Stack Hea 


gd 


Pe ees 
(Oo © 

oS 
CCS 


O ¢ 
la) 


iN CY CN CN! GN 
CO 


i<e) 


AMERICAN CONFERENCE 


OF GOVERNMENTAL 
INDUSTRIAL 


HYGIENISTS 


5-5 


5-6 Industrial Ventilation 


VELOCITY PRESSURE METHOD CALCULATION SHEET 


Problem #1 Class Designer Date 


Duct Segment Identification 


aah. = b-c e-f 
Target Volumetric Flow Rate cfm 390 1300 | 
Minimum Transport Velocity fpm a 
Maximum Duct Diameter inches 45 las | 
Selected Duct Diameter inches lg 45 las | 
Duct Area ft? 0.0873 0.1104 | 
Actual Duct Velocity fpm 
Duct Velocity Pressure "wg 


Maximum Slot Area 
Slot Area Selected ft? 


0.1104 


3631 


NI 


rfajleajeaiofearpajyarlaja]o 


N 
iF ~ 1 1 le | , 
on 


Slot Velocity 


Sot 
3 


Slot Velocity Pressure 


ae 
cers 
2 ata 
fe] 
Slot Loss Coefficient | 
ioe 2 
pe oN 
Eee 


AAOTrN 
= 
re} 


Acceleration Factor (0 or 1) 
Slot Loss per VP (13 + 14) 
Slot Static Pressure (12 x 15) 


Duct Entry Loss Coefficient (Fig. 5-13) on ey 
Acceleration Factor (1 or 0) Pin ee, 
Duct Entry Loss per VP (17 + 18) lies | 
[Ductentryloss B19) Wa Ls | Eee 

Se 
Hood Static Pressure (16 + 20 + 21) “wg 1.17 — 
lio 


a 

=e 

el 

ea 

traight Duct Length feet ee et 
0.0620 | 0.0620 | | 

ne 

fs 

nee 

ak | 


ZOo-aANCHM DOOLT 


NPN ED 
WEN 


TIED” 


riction Factor (H,) 0.0703 
riction Loss per VP (23 x 24) 


Number of 90 deg. Elbows 


on sl 
bow Loss Coefficient | | 
pee 
ee 


LB 
on 


Elbow Loss per VP (26 x 27) 
Number of Branch Entries (1 or 0) 
30 


Entry Loss Coefficient 


= 
et 
re 
—_ 
Lael 

ier Seamer 
Branch Entry Loss per VP (29 x 30) an 
Pia are 
Exec: a 
a 
ao 


Special Fitting Loss Coefficients 
33 | Duct Loss per VP (25 + 28 + 31 + 32) 
34 | Duct Loss (33 x 8) "wg 
Duct Segment Static Pressure Loss (22 + 34) “wg 20 
Other Losses (VP-VP,, etc. “wa Wares tal 
Cumulative Static Pressure "wg 536 
36 | Governing Sac Pressue ———SSSCS~“~*~*~“‘“‘*~*é~SdCCSC“‘(RWS Cd 


Corrected Velocity fpm 
Corrected Velocity Pressure “wg 
Resultant Velocity Pressure "wg 


Exhaust System Design Procedure 5-7 


Temperature Remarks: 


Elevation 


vp, =: yp, +2: vp, 
: 3 Q, 


Straight Duct Friction Loss 
ye 
nie = Qe 
[15 [ourmaeia [a | b_| 
16 


mm | 0.0425 0.602 | 
Flexible (fabric covered 0.604 0.639 
wires) 


Fan Static Pressure 
FAN SP = SP,,,. - SP, - VP iq 


Branch Entry Loss Coefficients 
Angle | Loss Coefficients 


90° Round Elbow Loss Coefficients 
(5 piece) 


60° elbow = 2/3 loss 
45° elbow = % loss 
30° elbow = 1/3 loss 


Adapted from Michigan Industrial 
Ventilation Conference (8/96) 


RL HRl[O]w 
-]!1/O/0O]o% 


42 


5-8 Industrial Ventilation 


Size for balance and 
transport velocity. 


Branch ducts 


TAPERED DUCT SYSTEM 


Maintains transport velocity 


Space collectors and fans to keep plenum 
size as small as practical. 


t 


Air enters fabric collectors 
through hopper. 

Separate duct for 
other types. 


To belt drive 
fon. 


Hopper of dry 
scollectors can 


Air lock if 
required. 


Size for 500 to 2000 fpm 
CT) Cy 


aa 


Cleanout door every 10° 


Branch ducts Branch ducts 


EXTENDED PLENUM SYSTEM 
Self cleaning type 


NOTE: Design plenum velocities are at the most 1/2 the branch duct 
design velocities and typically less than 2000 fpm. 


AMERICAN CONFERENCE 


PLENUM vs 
OF GOVERNMENTAL CONVENTIONAL SYSTEM | 
INDUSTRIAL HYGIENISTS a eae 


Exhaust System Design Procedure 5-9 


as plenum for 1500 — 2000 fpm. 


ae chain 


€ Soa al? Vall Na 
= } RO POON A TT 


cleaning main - drag chain 


Size plenum for 1500-7 


Be, § To collector — 
to 2000 fom. : ; 


and fan. | ee 
Deck plate - ed PPL Size for 


Corivenience 


oD 


3. Under floor ~ manual cleaning A. Large plenum ~ manual cleaning 


Plenum o 


~ Hopper iS . <a an 5 
“spy oe ~ v - fo collector 
lop per XX *, ; fin 
a — =i = eee ee as! 


Pneumatic cleaning duct. Size for - 
balance and transport velocity. 
with pneumatic cleaning 


Hopper duc 


L 
Reference 5.3 


NOTE: Design plenum velocities are at the most 1/2 tre branch duch 
design velocities and typically less than 2CO0O fpm. 
AME RICAN CONFERENCE 
OF G oe RNME NT ao 


TYPES OF PLENOMS 


5-10 Industrial Ventilation 


the fan (in terms of number of duct segments) and proceeds, 
segment by segment, to the fan. At each junction, the static 
pressure necessary to achieve desired flow in one stream must 
equal the static pressure in the joining air stream. The static 
pressures are balanced by suitable choice of duct sizes, elbow 
radii, etc., as detailed below. 


5.5.2 Blast Gate Method: The design procedure depends 
on the use of blast gates which must be adjusted after instal- 
lation to achieve the desired flow at each hood. At each 
junction, the flow rates of two joining ducts are achieved by 
blast gate adjustment which results in the desired static pres- 
sure balance. 


It isacommon practice to design systems on the assumption 
that only a fraction of the total number of hoods will be used 
at a time and the flow to the branches not used will be shut 
off with dampers. For tapered system designs, where particu- 
late is transported, this practice may lead to plugging in the 
main duct due to settled particulate. 


5.5.3 Choice of Methods: The Balance by Design 
Method is normally selected where highly toxic materials are 
controlled to safeguard against tampering with blast gates and 
consequently subjecting personnel to potentially excessive 
exposures. This method is mandatory where explosives, ra- 
dioactive dusts, and biologicals are exhausted because the 
possibility of accumulations in the system caused by a blast 
gate obstruction is eliminated. 


§.5.4 Balance by Design Procedure: The pressure loss 
of each duct segment is calculated from an exhaust hood to 
the junction with the next branch based on hood design data, 
fittings, and total duct length. At each junction, the SP for each 
parallel path of air flow must be the same. Where the ratio of 
the higher SP to the lower SP is greater than 1.2, redesign of 


the branch with the lower pressure loss should be considered. 
This may include a change of duct size, selection of different 
fittings, and/or modifications to the hood design. Where static 
pressures of parallel paths are unequal, balance can be ob- 
tained by increasing the air flow through the run with the 
lower resistance. This change in flow rate is calculated by 
noting that pressure losses vary with the velocity pressure and 
therefore as the square of the flow rate, so: 


SP, 
Qcorected = Qbesign SPaa 
uc! 


where the "governing" SP is the desired SP at the junction 
point and the "duct" SP is that calculated for the duct segment 
being designed. 


[5.1] 


5.5.5 Blast Gate Procedure: Data and calculations in- 
volved are the same as for the balanced design method except 
that the duct sizes, fittings, and flow rates are not adjusted; 
the blast gates are set after installation to provide the design 
flow rates. It should be noted that a change in any of the blast 
gate settings will change the flow rates in all of the other 
branches. Readjusting the blast gates during the system bal- 
ancing process sometimes can result in increases to the actual 
fan static pressure and increased fan power requirements. 


Recent work®? describes a method whereby blast gate 
settings can be made by means of pressure readings instead 
of by velocity readings. The biggest advantage of this method 
is that the process of resetting the insertion depths need not 
be a repetitive procedure. 


5.5.6 System Redesign: Many ventilation systems are 
changed after installation (processes are changed; operations 
are relocated; additional equipment is added to the production 
floor; etc.). When such changes occur, the effect of the pro- 


TABLE 5-1. Relative Advantages and Disadvantages of the Balance by Design Method and Blast Gate Method 
Blast Gate Method 


Balance by Design Method 


1. Flow rates cannot be changed easily by workers or at the whim of the 
operator. 


2. There is little degree of flexibility for future equipment changes or 
additions. The duct is "tailor made" for the job. 


3. The choice of exhaust flow rates for a new operation may be incorrect. 
In such cases, some duct revision may be necessary. 


4. No unusual erosion or accumulation problems will occur. 
5. Duct will not plug if velocities are chosen correctly. 


6. Total flow rate may be greater than design due to higher air 
requirements. 


7. The system must be installed exactly as designed, with all obstructions 
cleared and length of runs accurately determined, 


1. 


Flow rates may be changed relatively easily. Such changes are 
desirable where pickup of unnecessary quantities of material may affect 
the process. 


Depending on the fan and motor selected, there is somewhat greater 
flexibility for future changes or additions. 


. Correcting improperly estimated exhaust flow rates is relatively easy 


within certain ranges. 


Partially closed blast gates may cause erosion thereby changing 
resistance or causing particulate accumulation. 


Duct may plug if blast gate insertion depth has been adjusted 
improperly. 


Balance may be achieved with design flow rate; however, the net 
energy required may be greater than for the Balance by Design Method. 


Moderate variations in duct layout are possible. 


posed change(s) to the ventilation system should be calcu- 
lated. Often, systems are altered without adequate design, 
resulting in catastrophic changes to some hood flow rates. The 
result is that worker safety and health are jeopardized. 


5.6 AIDS TO CALCULATIONS 


As an alternative to performing these calculations manu- 
ally, programmable calculators and computers can be used to 
provide assistance with the design of systems. The ACGIH 
Industrial Ventilation Committee does not recommend any 
specific hardware or software. Many firms have developed 
their own software, and software packages are available com- 
mercially. Many of these software packages are available 
through ACGIH. 


5.7 PLENUM EXHAUST SYSTEMS 


Plenum systems differ from the designs illustrated earlier 
(see Figures 5-4 and 5-5). Minimum transport velocities are 
maintained only in the branch ducts to prevent settling of 
particulate matter; the main duct is oversized and velocities 
are allowed to decrease far below normal values, many times 
below 2000 fpm. The function of the main duct is to provide 
a low-pressure loss path for air flow from the various branches 
to the air cleaner or the fan. This helps to maintain balanced 
exhaust in all of the branches and often provides a minimum 
operating power. 


Advantages of the plenum-type exhaust system include: 


1. Branch ducts can be added, removed, or relocated at 
any convenient point along the main duct. 


2. Branch ducts can be closed off and the flow rate in the 
entire system reduced, provided minimum transport 
velocities are maintained in the remaining branches. 


3. The main duct can act as a primary separator (settling 
chamber) for large particulate matter and refuse mate- 
rial which might be undesirable in the air cleaner or 
fan. 


Limitations of this design include: 


|. Sticky, linty materials, such as buffing dust, tend to 
clog the main duct. It may be expected that greatest 
difficulty will be encountered with the drag chain type 
of cleaning, but the other types will be susceptible to 
buildup as well. 


2. Materials that are subject to direct or spontaneous 
combustion must be handled with care. Wood dust has 
been handled successfully in systems of this type; 
buffing dust and lint are subject to this limitation and 
are not recommended. Explosive dusts such as mag- 
nesium, aluminum, titanium, or grain dusts should not 
be handled in systems of this type. 


5.7.1 Choice of Systems: Various types of plenum ex- 
haust systems are used in industry (see Figure 5-5). They 


Exhaust System Design Procedure 5-11 


include both self-cleaning and manual-cleaning designs. Self- 
cleaning types include pear-shaped designs which incorporate 
a drag chain conveyor in the bottom of the duct to convey the 
dust to a chute, tote box, or enclosure for disposal. Another 
self-cleaning design uses a rectangular main with a belt con- 
veyor. In these types, the conveyors may be run continuously 
or on periodic cycles to empty the main duct before consid- 
erable buildup and clogging occur. A third typeS*) of self- 
cleaning design utilizes a standard conveying main duct 
system to remove the collected material from a hopper-type 
of main duct above. Such a system is usually run continuously 
to avoid clogging of the pneumatic air circuit. Manual-clean- 
ing designs may be built into the floor or may be large 
enclosures behind the equipment to be ventilated. Experience 
indicates that these should be generously oversized, particu- 
larly the underfloor designs, to permit added future exhaust 
capacity as well as convenient housekeeping intervals. 


5.7.2 Design: Control flow rates, hoods, and duct sizes for 
all branches are calculated in the same manner as with tapered 
duct systems. The branch segment with the greatest pressure 
loss will govern the static pressure required in the main duct. 
Other branches will be designed to operate at this static 
pressure or locking dampers can be used to adjust their 
pressure loss to the same static pressure as the governing 
branch. Where the main duct is relatively short or where the 
air cleaners or fans can be spaced along the duct, static 
pressure losses due to air flow in the main duct can be ignored. 
For extremely long ducts, it is necessary to calculate the static 
pressure loss along the main in a manner similar to that used 
in the balanced and blast gate methods. Design plenum ve- 
locities are at most one-half the branch velocity design duct 
velocities and typically less than 2000 fpm. Duct connections 
to air cleaners, fans, and discharge to outdoors are handled in 
the normal manner. 


5.8 FAN PRESSURE CALCULATIONS 


Exhaust system calculations are based on static pressure; 
that is, all hood static pressures and balancing or governing 
pressures at the duct junctions are given as static pressures 
which can be measured directly as described in Chapter 9. 
Most fan rating tables are based on Fan Static Pressure. An 
additional calculation is required to determine Fan Static 
Pressure before selecting the fan. 


5.8.1 Fan Total Pressure (FTP) is the increase in total 
pressure through or across the fan and can be expressed by 
the equation: 


FTP = TPoutlet — TPiniet [5.2] 


Some fan manufacturers base catalog ratings on Fan Total 
Pressure. To select a fan on this basis the Fan Total Pressure 
is calculated noting that TP = SP + VP: 


FTP = (SPoutlet + VPoutlet) — (SPintet + VPintet) [5.3] 


§-12 Industrial Ventilation 


5.8.2 Fan Static Pressure: The Air Movement and Con- 
trol Association Test Code defines the Fan Static Pressure 
(FSP) as follows: "the static pressure of the fan is the total 
pressure diminished by the fan velocity pressure. The fan 
velocity pressure is defined as the pressure corresponding to 
the air velocity at the fan outlet."°* Fan Static Pressure can 
be expressed by the equation: 


FSP = FTP — VPiniet [5.4] 
or 
FSP = SPouttet — SPintet — VPinlet [5.5] 


In selecting a fan from catalog ratings, the rating tables 
should be examined to determine whether they are based on 
Fan Static Pressure or Fan Total Pressure. Fan system effects 
(see Chapter 6) should also be considered when selecting a 
fan. The proper pressure rating can then be calculated keeping 
in mind the proper algebraic signs; i.e., VP is always positive 
(+), SPintet is usually negative (—), and SPoute is usually 
positive (+). 


5.8.3 Completion of the Example on Figure 5-3: To de- 
termine the Fan Static Pressure and Fan Total Pressure, note 
that the second column adds the fabric pressure drop through 
the bags in the collector. Column 3 adds the losses from the 
clean air plenum to the fan inlet, and the last column deter- 
mines the pressure losses through the stack. 


The FSP and FTP can be calculated from these values. At 
the outlet of the fan, the SP must be 0.48 "wg. At the inlet to 
the fan, the SP is -6.56 "wg. The VP at both locations is 0.78 
"we. From Equation 5.3, the system FTP = (0.48 + 0.78) — 
(-6.56 + 0.78) = 7.04 "wg. From Equation 5.5, the FSP = 0.48 
— (-6.56) — 0.78 = 6.26 "wg. 


5.9 CORRECTIONS FOR VELOCITY CHANGES 


Variations in duct velocity occur at many locations in 
exhaust systems because of necessary limitations of available 
standard duct sizes (area) or due to duct selections based on 
balanced system design. As noted earlier, small accelerations 
and decelerations are usually compensated automatically in 
the system where good design practices and proper fittings are 
used. There are times, however, when special circumstances 
require the designer to have a knowledge of the energy losses 
and regains which occur since these may work to his advantage 
or disadvantage in the final performance of the system. 


5.9.1 Branch Entries to Main Ducts: Sometimes the fi- 
nal main duct velocity exceeds the higher of the two velocities 
in the branches entering the main. If the difference is signifi- 
cant, additional static pressure is required to produce the 
increased velocity. A difference of 0.10 "wg or greater be- 
tween the main VP and the resultant VP of the two branches 
should be corrected. 


At any junction point, energy must be conserved. The 


energy entering each of the two air streams would be Q(TP) 
= Q(SP+VP). The first law of thermodynamics states that the 
sum of these must equal the energy leaving, or 


Q1(VP1+SP1) + Q2(VP2+SP2) = Q3(VP3+SP3) + Losses 


Note that the overall losses would be: 
Losses = F,Q,VP, + FoQ VP, 


where the subscripts refer to the ducts shown in Figure 5-6. 
In this manual, F, is considered to be zero and F, is given on 
Figure 5-15. Assuming we are balanced and the junction 
losses are included such that SP, = SP, and Q, = Q, + Q, (see 
Figure 5-6), there might be an additional change in static 
pressure due to the acceleration or deceleration of the gas 
stream. The following equation shows this effect: 


Q Q 
SP, + VP; = SP, +] — |VP, +| 2 \VP. 
3 3 4 [2 ve, Gl 2 


The last two terms on the right are defined as the resultant 
velocity pressure, VP, ; this can be simplified to 


_| Q Q, 
VP. = Cae + Gag [5.6] 


where: 
VP, = resultant velocity pressure of the combined 
branches 
Q, = flow rate in branch #1 
Q, = flow rate in branch #2 
Q3 = combined flow rate leaving the junction 


Note that the above equation is valid for all conditions, 
including merging different density gas streams, as long as 
the velocity pressures include the density effects. Also note 
that, if the flow rate through one branch was changed to 
balance at the branch entry, the velocity pressure and cor- 
rected flow rates should be used in Equation 5.6. 


The resultant velocity pressure (VP,) is computed using 
Equation 5.6. If VP3 is less than VP, , a deceleration has 
occurred and SP has increased. If VP3 is greater than VP, , an 
acceleration has occurred, and the difference between VP; and 
VP, is the necessary loss in SP required to produce the increase 
in kinetic energy between VP; and VP, . The correction is 
made as follows: 


SP3 = SP; — (VP3— VP;) [5.7] 
where: 
SP3 = SP in main #3 
SP, = SP at branch #1 = SP at branch #2 
VP3 
It should be noted that many designers believe a conserva- 


tive approach to fan selection would be to ignore any correc- 
tion if VP, is larger than VP3. 


velocity pressure in main #3 


A simpler equation for VP, was used in prior editions of 
this manual: 


2 
vp =[|_+ 22 

4005(A, + A.) 
This equation gives acceptable results (less than a 4% error) 


when the velocities of the two merging air streams are within 
500 fpm of each other. 


EXAMPLE 
: 6]. [6 |+ 
Lo | SD) 
Ue a 
| duct No | dio. | areo| a | v | we | se | 
(i) 10 | 0.545 | ble 0.79 2 
(2) 4 | 0.087} 340 | 3890 | 0.94 | ~2.11| 
| Main (3) | 10 | 0.545] 2275] 41701 1.08 | ~ | 
FIGURE 5-6. Branch entry velocity correction 
With the data shown, 
94 
p _ (1935)(0.79) , (340)(0.94) _ 9 a, “ws 


‘ 2275 2275 


SP = SP, —(VP3 —- VP.) =—2.11—(108 — 0.81) 
=-211-0.27=-2.38"wg 


Therefore, in this situation, an additional -0.27 "wg should 
be added to the junction SP to account for losses in pressure 
due to acceleration of the air stream. 


5.9.2 Contractions and Expansions: Contractions are 
used when the size of the duct must be reduced to fit into tight 
places, to fit equipment, or to provide a high discharge veloc- 
ity at the end of the stack. Expansions are used to fit a 
particular piece of equipment or to reduce the energy con- 
sumed in the system by reducing velocity and friction. Expan- 
sions are not desirable in transport systems since the duct 
velocity may become less than the minimum transport veloc- 
ity and material may settle in the ducts. 


Regain of pressure in a duct system is possible because 
static pressure and velocity pressure are mutually convertible. 
This conversion is accompanied by some energy loss. The 
amount of this loss is a function of the geometry of the 
transition piece (the more abrupt the change in velocity, the 
greater the loss) and depends on whether air is accelerated or 
decelerated. Loss is expressed as a loss coefficient multiplied 
by the velocity pressure in the smaller area duct of the transi- 
tion piece. One minus the loss factor is the efficiency of the 
energy conversion or regain. 


Exhaust System Design Procedure 5-13 


A perfect (no loss) contraction or expansion would cause 
no change in the total pressure in the duct. There would be an 
increase or decrease in static pressure corresponding exactly 
to the decrease or increase in velocity pressure of the air. In 
practice, the contraction or expansion will not be perfect, and 
there will be a change in total pressure (see Figure 5-7). In 
each example, total pressure and static pressure are plotted to 
show their relationship at various points in each system. See 
Figure 5-16 for design data. 


5.10 SAMPLE SYSTEM DESIGN 


A discussion of the calculations for either tapered duct 
method can best be done by a typical example using the 
exhaust system shown in Figure 5-8. Calculation sheets illus- 
trate the orderly and concise arrangement of data and calcu- 
lations (see Figures 5-9 and 5-10). The procedure outlined in 
Section 5.3 was used to develop the design. Each column is 
for a constant diameter duct segment that starts at a hood, 
junction point, air cleaning device, fan, or transition point. 


The problem considered is a foundry sand-handling and 
shake-out system. A minimum conveying velocity of 3500 
fpm is used throughout the problem except in ducts where 
excess moisture or dust loading increases that value. The 
operations, hood designations on the diagram, VS-print ref- 
erences, and required flow rates are presented in Table 5-2. 


5.11 DIFFERENT DUCT MATERIAL FRICTION LOSSES 


The friction loss table, Table 5-5, provides average values 
for galvanized sheet metal duct material (0.0005 feet equiva- 
lent sand grain roughness, where the roughness height repre- 
sents the average height of the roughness elements of the 
material). Table 5-6 provides the same information for black 
iron and other materials possessing a roughness height of 
0.00015 feet. Recent research indicates that an equivalent 
sand grain roughness factor of 0.0003 feet more accurately 
reflects the losses incurred in new HVAC galvanized duct 
systems. However, past experiences in industrial ventilation 
applications successfully reinforce the application of the 
0.0005 feet equivalent sand grain roughness factor. This may 
be due to inherently shorter duct runs and dustier environ- 
ments which are common within industrial ventilation appli- 
cations. The values in both tables can be used with no 
significant error for the majority of designs but special con- 
siderations may be desired if environmental conditions could 
significantly affect the duct design parameters. If the design 
requires special material, operates at a non-standard density, 
or is very hot, the duct material manufacturer should be 
consulted for the anticipated friction loss. 


5.12 FRICTION LOSS FOR NON-CIRCULAR DUCTS 


Round ducts are preferred for industrial exhaust systems 
because they provide a more uniform air velocity to resist 
settling of material and an ability to withstand higher static 


5-14 


industrial Ventilation 


EXAMPLE 1 — DUCT LOCATED ON SUCTION SIDE 
OF FAN 


Velocity changes as indicated. Since all the duct is on 
the suction side of the fan, TP at the fan inlet (point F) is 
equal to VP at the fan inlet plus the total duct resistance up 
to that point. This equals -4.2 SP since static pressure on 
the suction side of the fan is always negative. The duct 
system is the same as was used in Example 2 and therefore 
has the same overall resistance of 3.2. If it is again as- 
sumed that the inlet and discharge of the fan are equal 
areas, the total pressure across the fan will be the same as 
in Example 2 and, in each case, the fan will deliver the same 
air horsepower when handling equal volumes of air. 


Static pressure conversion between B and C follows 
contraction formula (Figure 5-16). There must be sufficient 
SP at B to furnish the additional VP required at C. In 
addition, the energy transfer between these two points is 
accompanied by a loss of 0.3. Since SP at B =—2, SP atC 
= —~2.0 + (-1.0) + (-0.3) = -3.3 “wg. 


Static pressure regain between D and E follows the 
regain formulae (Figure 5-16). If there were no loss in the 
transition piece, the difference of 1 “wg in velocity pressure 
would be regained as static pressure at E, and SP at that 
point would be -2.8. However, the transition is only 60% 
efficient (0.4 loss) so the SP at E =—2.8 + (-0.4) = -3.2. 


30% !oss 


1) mete 


ransscomeaceares 
Aimospheric pres 


EXAMPLE 2 — DUCT LOCATED ON DISCHARGE 
SIDE OF THE FAN 


Velocity changes as indicated. The duct is located on 
the discharge side of the fan. Total pressure at the fan 
discharge (point A) is equal to the velocity pressure at the 
discharge end of the duct (point F) plus the accumulated 
resistances. These add up to 1.0+1.0+0.4+05+0.3 
+1.0= 4.2. 


Static pressure regain between D and E follows the 
regain formulae (Figure 5-16). If there were no energy 
loss in the transition piece, static pressure at D would be 
O because the difference in VP of 1 would show up as 
static pressure regain. However, the transition is only 
60% efficient which means a loss of 0.4, so SP at point D 
=0+04=04. 


Conversion of static pressure into velocity pressure 
between B and C follows contraction formulae (Figure 
5-16). There must be sufficient static pressure at B to 
furnish the additional velocity pressure required at C. In 
addition, transformation of energy between these two 
points is accompanied by a loss of 0.3. Since SP at C = 
0.9, SP atB =0.9+0.3+ 1.0 = 2.2. Since there is no duct 
onthe suction side of the fan, total pressure against which 
the fan is operating is 4.2". 


EXPANSIONS 
AND CONTRACTIONS 


[cum on = : 


TABLE 5-~2. Details of Operation 


Minimum 
No. Hood No. VS-Print Exhaust, cfm 
1. Vibrating Shakeout 1 20-02 9600 
4' x 6' grate 
2. Shakeout hopper 2 20-03 960 
3. Vibrating pan feeder 3 20-03 700 
24" wide 
4. Incline sand belt 5 700 
24" x 28" long 
5. Magnetic pulley 
6. Tramp iron box 
7. Bucket elevator 7a(lower) 50-01 250 
24" x 30" casing 7b(upper) 250 
8. Vibrating screen 24 ft? 8 99-01 4200 
9. Sand bin 600 ft° 9 50-10 500 
18" x 20" opening 
10. Waster sand box 44" x 10 99-03 1225 
54", 6" clearance (V = 150 fpm) 
11. Sand weigh hopper 11 60-02 900 
12. Sand muller 6' dia. 12 60-02 
13. Wet dust collector 
(includes fan) 
DIMENSIONS 
No. of CFM 
Branch Required, Straight 
or Main Minimum Run, ft Elbows Entries 
1-A 9600 13 1-90° 
2-B 960 3 1-60° 1-30° 
3-B 700 4 4-90°+1-60° 1-30° 
B-A 1660 18 2-90° 1-30° 
A-C 11,260 34 
5-D 700 7 1-30°+1-60° 1-30° 
7a-D 250 5 
D-C 950 14 1-90°+1-60° 1-30° 
C-E 12,210 6.5 
8-F 1200 11 2-90° 
9-F 500 4 1-90°+1-60° 1-30° 
F-G 1700 5 
70-G 250 15 1-60° 1-30° 
G-E 1950 6 1-60° 1-30° 
E-H 14,160 35 
10-J 1225 6 1-45° 
12-J 900 2.5 1-30° 1-30° 
J-H 2125 8 1-90°+1-60° 1-30° 
H-K 16,285 9 2-45° 
13 16,285 
14-L 16,285 20 


Exhaust System Design Procedure 5-15 


pressure. At times, however, the designer must use other duct 
shapes. 


Rectangular duct friction can be calculated by using Table 
5-5 or 5-6 in conjunction with Table 5-9 to obtain rectangular 
equivalents for circular ducts on the basis of equal friction 
loss. It should be noted that, on this basis, the area of the 
rectangular duct will be larger than the equivalent round duct; 
consequently, the actual air velocity in the duct will be re- 
duced. Therefore, it is necessary to use care to maintain 
minimum transport velocities. 


Occasionally, the designer will find it necessary to estimate 
the air handling ability of odd-shaped ducts. The following 
procedure®*) will be helpful in determining the frictional 
pressure losses for such ducts. The wetted perimeter in the 
following discussion is the inside perimeter of the odd-shaped 
duct corresponding to the cross-sectional area. 


1. Find duct cross-sectional area, ft? 2... ...20202. A 
2. Find wetted perimeter, ft ...........2., P 
3. Calculate hydraulic radius, ft ...... R (R= A/P) 
4. ConvertR toinches............ r(r= 12R) 
5. Calculate equivalent diameter, in... .. D (D = 4r) 
6. Use the proper friction table based on the equivalent 


diameter and flow rate (or velocity). 
5.13 CORRECTIONS FOR NONSTANDARD DENSITY 


Fan tables and exhaust flow rate requirements assume a 
standard air density of 0.075 Ibm/ft?, which corresponds to 
sea level pressure, no moisture, and 70 F. Changes in air 
density can come from several factors including elevation, 
temperature, internal duct pressure, changes in apparent molecu- 
lar weight (moisture content, gas stream constituents, etc.), and 
amount of suspended particulate. Where appreciable variation 
occurs, the change in air density must be considered. 


Factors for different temperatures and elevations are listed 
in Table 5-10. Correction for temperatures between 40 F and 
100 F and/or elevations between —1!,000 feet and +1,000 feet 
are seldom required with the permissible variations in usual 
exhaust system design. 


Similarly, if internal duct pressures vary by more than 20 
"weg from standard pressure, the density will change by over 
5%. If there is excessive moisture in the airstream, the density 
will decrease. Suspended particulate is assumed to be only a 
trace impurity in industrial exhaust systems. If there are 
significant quantities of particulate in the duct system, this 
addition to the air stream density should be addressed. This 
field is called material conveying and is beyond the scope of 
this manual. 


Many times, the system designer is confronted with a 
combination of these five means of changing density. If so, 
then the density of the air stream should be determined and, 
if the density is more than 5% different from standard density, 


Industrial Ventilation 


5-16 


Plan View 


“-—— 
fa) 
Oo 

2 
a 
w 
~S 
oO 
io) 
N 
tl 
No 
2 
aS) 
3 
— 
oJ 
| 
n 
= 
re) 
2 
o 
<x 


All duct lengths ore ¢ to ¢ 


Branch entries — 30 


Elevation View 


| AMERICAN CONFERENCE | 
| OF GOVERNMENTAL | 


PROBLEM 2 


Exhaust System Design Procedure 5-17 


VELOCITY PRESSURE METHOD CALCULATION SHEET 


Problem #2 Class Designer Date 


Duct Segment Identification 4-A 


rn 
(ve) 


wn 
o 
@ 
° 
Q 
® 
Q 
Oo 
< 
a 
2 
Oo 
i 
3 
® 
Q 
© 
Q 
5 
7 fe) 
? > 
® 
n 
on 
on 


Ey aa 

Target Volumetric Flow Rate cfm fo60 L700 ats | | 9600 _| 
Minimum Transport Velocity fpm 4000 | 4000 _| 
Maximum Duct Diameter inches 20.97 fT 
5 |e {20 

Duct Area fe | 1.7671 _| 
Actual Duct Velocity fpm 4400 | s432__| 


7 


Duct Velocity Pressure wg 1.21 
ft? 


4.8 
4.8 
2000 


Slot Area Selected ft? 


ia 
Weed 
Ss Slot Velocity fpm ee. bc Al 
i sal 
il 


= 
= 


H |} L | Slot Velocity Pressure “wg 0.25 0.25 
13 a - Slot Loss Coefficient 1.78 1.78 
D |S | Acceleration Factor (0 or 1) aa) 
15 | 5 Slot Loss per VP (13 + 14) al 
16 |U Slot Static Pressure (12 x 15) Hie i 0.44 
Cc — os 
T | Duct Entry Loss Coefficient (Fig. 5-13) 25 
J Acceleration Factor (1 or 0) a as 
9 | N | Duct Entry Loss per VP (17 + 18) 41.25 1.25 


Duct Entry Loss (Bx 19) ‘wa 


Other Losses : 


wo] | 
Hood Static Pressure (16 + 20 + 21) wg 1.40 


Straight Duct Length fee 

Friction Factor_(H)) 0299 0098 

Number of 90 deg. Elbows 41.67 2 
19 : 


2.30 


2.74 


NENT NFS 
NTS] oO 


~~ 


0110 
14 


lag7 l2 | 

Elbow Loss Coefficient re yee é 
Elbow Loss per VP (26x27) 
Number of Branch Entries (ort) |, sofa 
Entry Loss Coefficient 
Branch Entry Loss per VP (29 x 30) 
Special Fitting Loss Coefficients law 

Las | 


Duct Loss per VP (25 + 28 + 31 + 32) 


33 
3 


lia | 
e600 __| 
| 4000__| 
[20.97 | 
[20 
[2.1817 | 
4400 _| 
ge | 
| 2000 __| 
lo25s | 
ron 
Die 
1.78 
Loss | 
L425 | 
[151 
Nese | 
[495 
Se aa 
[0098 


No 
a 


(ev) 


2 
33 


_— 
= 
oO 


Duct Loss (33 x 8) 


CO 
~ 


| 33 

= 

Duct Segment Static Pressure Loss (22 + 34) “wg -2.18 -2.34 -3.61 
Other Losses (VP-VP,, etc. "wa ee eee ee 
Cumulative. Static Pressure l-218 | 
Governing Static Pressure “wg 
| 42 | 


42 


a 


-2.34 -3.61 
~4.25 
10416 
5894 


2.16 
2.07 


Corrected Volumetric Flow Rate cfm 
Corrected Velocity fpm 


Corrected Velocity Pressure "wg 


Resultant Velocity Pressure zi 


FIGURE 5-9. Balanced design method 


5-18 Industrial Ventiiation 


Temperature Remarks: 


Elevation 


13130 
4000 
24.53 


Pertinent Information 
From Chapter 5 


Pyov 


ae 


3500 
6.05 


3500 3500 


[700 | 
| 3500 
| 07 
0.1650 _| 


3500 


NO 
aSS 


Qoor = = Qgesign 


0.1963 
3565 
0.79 


0.1650 0.2673 0.1963 


4864 


3.1416 
4179 
1.09 


3742 
0.87 


3573 
0.80 


1.47 Q, Q, 
akon VP,+ a, VP, 


VP. = 
Straight Duct Friction Loss 
ve 
Hf=a [oud 
0612 


f Black iron, Aluminum, 
PVC, Stainless steel ees ne0e 


9 


1.25 1.25 


1.40 


Es 


0478 


— 
Dp 


0436 
0 


0361 0425 0079 


D-C 
955 g 
7 
14 
5 
4 467 1.67 
9 _|49 | 49 
9 32 ; 
1 
18 
18 
1.01 
2.99 


[oe | 
[13130 _| 
| 4000 _| 
[24.63 | 
[2a 
[aiaie_| 
|4i7o | 
[1.09 | 
aa 
Pee 
Mica aaa 
Ee ad 
at 
eeu 
a 
ae 
a 
bn 
beset 
ae 
cee] 
ees] 
[0079 _| 
ae ded 
Ze eel 
a 
chee eae! 
re 
cae ees 
= eae 
Los 
Log 
|o6 | 
aed 
eae 
7 
a 
= a 


re | 
re] 
a 
| 
ra 
re | 
Par | 
rae | 
Par | 
Es 
: 
rar | 


(a) 


1.10 
1.61 
1.61 


[ss | 
aa 
| o47e | 
haa 
Lig 
ea 
a fis | 
as 2 
aa 
eta 
za 
[219 | 


D 
00 
5 
25 
3 33 
1 
1 
i 19 
it 1 
J 18 
18 18 
& 70 
5 79 
1 2.19 
1.06 


1.52 2.19 


ae 
al ore E 


D- 
5 
48 

1 
3 

j 

4 

sf 

4 


5- 
Z 
5 
1 
7 
1 


7 
3 
5 


4.21 


{o) 


C 

5 
9 
2. 
8 
8 
4 


1003 
5109 


re EN 
BR fea 
iad rs 


S 


FIGURE 5-9. Balanced design method (continued) 


Flexible (fabric covered } 9 9, 86 | 0.604 
wires) 


0.639 


Fan Static Pressure 
FAN SP = SP, - SPi, - VPin 


Branch Entry Loss Coefficients 


90° Round Elbow Loss Coefficients 
(5 piece) 


60° elbow = 2/3 loss 
45° elbow = 1/2 loss 
30° elbow = 1/3 loss 


Adapted from Michigan Industrial 
Ventilation Conference (8/96) 


Exhaust System Design Procedure 5-19 


VELOCITY PRESSURE METHOD CALCULATION SHEET 


Problem #2 (Continued) Class Designer Date 


Duct Segment Identification 7b-G 
Target Volumetric Flow Rate cfm 250 
Minimum Transport Velocity fpm 3500 
Maximum Duct Diameter inches i703 [5 | dose _| 3.62 
Selected Duct Diameter inches an ee Poe Pee 3.5 
Duct Area ft* | 0.2673 0.0668 
Actual Duct Velocity fpm 3742 
Duct Velocity Pressure “wg 
Maximum Slot Area ae 


Slot Area Selected 


— 


= 


7 


ft 
ft 


— 
= 


2 
2 
m 
g 


NAOTMN 


aera Ee 
aioe (een 
a ees ee 
et ae ee 
eae a ee 
Acceleration Factor (0 or 1) ae a ee 
is ae ee 
en aa eae 


Duct Entry Loss Coefficient (Fig. 5-13) 
[| Acceleration Factor tr gg Sd 
N [Ductentytossperve (#18) dts tts tas | sd 
|DuctEntyLoss X19) wtp tos ttn =| Ss za 
Other Losses “wg ar reas ere eee 
Hood Static Pressure (16 + 20 + 21) “wg fiso_|105 |ico | saya _| 
5 7 


—-ANCH DOOT 
a 
2 
~~ 
° 
n 
n 
no] 
2 
< 
U 
os 
w 
+ 
_—sS 
+= 


> 


Ww ow BIWILWIWIWEININI NIN MIME LM ol al] fapa eis rei 
BRPWINM O;fo|]a oO NlAlLOLO,O NED! aA; HR] WEN 


O | Corrected Velocity fpm 


41 | Corrected Velocity Pressure “wg 


at 
20 | Comected Voumetic lowRale SSCS 
neal 
ar 


j131 [se | 


lu 4 

4 
25 
Number of 90 deg, Elbows 2 ae? tar? 

7 [Elbow Loss Coefficient id tg tt tg OT dt 
[ElbowLosspervP (26x27) tc Ce te | dt 
[Number ofBranchEntries tort) | tg St ST 
[Entry Loss Coefficient Tt tt | dt 

1_|BranchEntrytosspervP 29x30) ct ct St 
Special Fitting Loss Coefficients a ae ee eee eee 

Duct Loss (33 x 8) ‘wolor [eo [os [az  [as7___ 

35 | Duct Segment Static Pressure Loss (22 + 34) “wg 
6 | Other Losses (VP-VP,, etc. C77 Re) er 
37 | Cumulative Static Pressure "wg 
8 | Governing Static Pressure "wg ee ee ec 
| e309 aver 

re 

eee Ee een 

ee [se | 


42 | Resultant Velocity Pressure "wg | 1.31 


FIGURE 5-9. Balanced design method (continued) 


5-20 Industrial Ventilation 


Temperature 


Elevation 


G-E 
2017 
3500 
10.28 


= 
Oo 


0.5454 
3698 
0.85 


ua 
(eee (co) 


= jo 
foomm ice] 


wo tp 
co 1 


[oe |. 
[2017 | 
| 3500__| 
Lio28 _| 
Mion] 
losasa | o. 
| 3608 
loss | 4. 
aes 
an 
ase 
sal 
a 
pened 
[ 
eae 
aes 
ee 
Meee a 
| 
ae 
Paar 
rae 
| o2s2 |. 
fsa | 
ar ls 
Lio 
Roa 
cree 
i ae 
oar 
Las 


38 


FIGURE 5-9. Balanced design method (continued) 


Remarks: 


gov 
SPauct 
5147 
1.65 


_Q, jase 


2 
se a, YE: 


Straight Duct Friction Loss 
ye 
Hy = a oe 
2p [| 
0.533 | 


Black iron, Aluminum, 
PVC, Stainless steel 0.465 
ej eet | Gosty | 0.604! | 20.639 


Fan Static Pressure 
FAN SP = SP,,, - SP;, - VPin 


90° Round Elbow Loss Coefficients 
(5 piece) 


60° elbow = 2/3 loss 
45° elbow = 1/2 loss 
30° elbow = 1/3 loss 


Adapted from Michigan Industrial 
Ventilation Conference (8/96) 


Exhaust System Design Procedure 5-21 


VELOCITY PRESSURE METHOD CALCULATION SHEET 


Problem #2 (Continued) Class Designer Date 


K-FAN 


= 


Duct Segment Identification 


17800 
2600 | 2600 _| 
aes 
[sa 


Target Volumetric Flow Rate cfm 
Minimum Transport Velocity fpm 
Maximum Duct Diameter inches 
Selected Duct Diameter inches 


Duct Area ft? 6.3050 


“I 


2590 


Actual Duct Velocity fpm 


nN 
o> 
ko 
(o 
m) 
or 
i 


= 
a 
Fiecaes Poveeh feeeaee ee eae 
i co 


Other Losses 


Hood Static Pressure (16 + 20 + 21) 
6 


aQ 


wd 


2 
0051 0053 


o1 


n 


traight Duct Length feet 


n 


34 

Duct Velocity Pressure "wg | 1.45 0.42 
Maximum Slot Area ft? eens 
Slot Area Selected ft? Mee 
Ss Slot Velocity fpm aan 
H | L | Slot Velocity Pressure "wg ee 

obey 
ott Slot Loss Coefficient Ee 
4 D {|S | Acceleration Factor (0 or 1) es 2 
es Slot Loss per VP (13 + 14) a 
U Slot Static Pressure (12 x 15) fF 
: Duct Entry Loss Coefficient (Fig. 5-13) a 
‘ Acceleration Factor (1 or 0) eeee = | 
ny | Duct Entry Loss per VP (17 + 18) ae 
Duct Entry Loss (8 x 19) "wg i 


riction Factor (H,) 0071 
Friction Loss per VP (23 x 24) 


| 26 | Number of 90 deg. Elbows 
27 | Elbow Loss Coefficient 1 


21 
2 


Number of Branch Entries (1 or 0) 


Entry Loss Coefficient 


Branch Entry Loss per VP (29 x 30) 
pecial Fitting Loss Coefficients 
uct Loss per VP (25 + 28 + 31+ 32) 25 

34 (33 x 8) 
uct Segment Static Pressure Loss (22 + 34) 


Other Losses (VP-VP,, etc. 


Oln 
oO 
here 
ua 
raat 


is) 
= 
° 
ss 
a 
ie) 
a 
a 
ee ee oe one 
w in 


oO 
¢lele 
Go |a 
No 
cn 


= 
ico} 
eo 
N 


= 
on 


4 c ph i> 
on jon 


tes) 


eae PTT 
Corrected Velocity fpm 
Corrected Velocity Pressure "wg 
Resultant Velocity Pressure "wg 


FIGURE 5-9. Balanced design method (continued) 


5-22 


Industrial Ventilation 


NOTES 


1. 


Balancing at B: SP ratio = -2.51/-2.18 = 1.15, so the 
flow rate through the lower resistance run can be 
corrected. From Equation 5.1, 


—2.51 
Qeorrected = 700 218 =751 cfm 


From Equation 5.6, 


= 269 449491 429 -140"wg 

1711 1711 
Note that the numbers for 3-B reflected the corrected 
values. The velocity pressure of 1.29 "wg corre- 
sponds to the new velocity after correcting the flow 
rate to 751 cfm (751/0.165 = 4551 fpm, up from 4243 
fpm). 
The VP in duct B-A is 1.50, while the VP; at A was 
1.40, so there was an acceleration at B of 0.10. This 
is reflected in the overall pressure drop to point A 
(2.51 + 0.10 + 1.64 = 4.25 "wg.) Column 19 is a 
convenient place to enter this additional SP drop. 


Balancing at A: Initial SP ratio at A (-4.25/-2.34 = 
1.82) is too high to allow flow rate increase, so 1-A 
was redesigned with an 18" dia. duct to result in an 
acceptable SP ratio (4.25 + -3.61 = 1.18). Then 


VP, 


Qeorrected = 9600 == = 10416 cfm 


From Equation 5.6, VP; = 2.07 "wg. As the VP in the 


FIGURE 5-9. Balanced design method (continued) 


segment from A to C was clearly less than VP; , no 
correction to the SP was made. 


This same procedure was followed at each of the 
other junction points until reaching the collector. At 
junction points J and H, the SP ratio was slightly over 
the 1.20 recommended. However, the flows were 
increased using Equation 5.1 anyway because re- 
ducing the diameter would have resulted in unac- 
ceptably high duct velocities and SP drops. 


K is the inlet to that collector and the fan is labeled 
FAN. Assuming the collector loss of 4.5 "wg is the 
"flange-to-flange" loss, and 2 feet of straight duct 
separates the fan from the collector, it might be 
advisable to use a non-standard diameter duct equal 
to the fan's inlet to connect the two devices. That is 
why a 35.5" diameter was chosen here. 


The ductloss calculations for the stack (FAN-L) show 
an overall loss of 0.05 "wg. However, there is an 
additional effect to consider. Most centrifugal fans 
have an exit area virtually the same as the inlet area. 
If so, then there would be an additional acceleration 
from approximately 0.42 "wg to the duct VP of 0.50 
"wg, or 0.08 "wg. This acceleration "loss" is included 
in line 19 of the FAN-L column. 


The above FSP calculation assumes that there are 
no losses due to the contraction from the fan exit to 
the 32" diameter duct. This is usually a small enough 
loss to ignore. For instance, if the contraction half- 
angle was 15 degrees with a loss factor of 0.08 (see 
Figure 5-16), the maximum error in ignoring this loss 
would be less than 0.02 "wg. 


Exhaust System Design Procedure 5-23 


VELOCITY PRESSURE METHOD CALCULATION SHEET 


Problem #2 With Blast Gates_ Class Designer 


Duct Segment Identification 4 2- A-C 


3B 
Target Volumetric Flow Rate cfm lo600 | o6o | 700 4660 11260 
Minimum Transport Velocity fpm 4000 4000 
aximum Duct Diameter inches 5.66 22.72 
Selected Duct Diameter inches lon ld 55 Pee 4 22 
ft? | 2.1917 0.1650 2.6398 

Actual Duct Velocity fpm 4243 4265 
Duct Velocity Pressure wg 1.12 113 
Maximum Slot Area fe te 25) 


lot Area Selected ft? 


| 


oO 
Cc 
Q 
Q 
> 
a 
oO 
© 


lot Velocity 


no 


ot Velocity Pressure wg 


ot Loss Coefficient 


a 
ae 
ee 
| 
cceleration Factor | | 
ae 
Lee 


> 


_ 
b 
nNAAOMFNn 


= 
a 
icp] 


ot Loss per VP (13 + 14) 

ot Static Pressure (12 x 15) ; 

Duct Entry Loss Coefficient (Fig. 5-13) 5 
Acceleration Factor 

Duct Entry Loss per VP (17 + 18) 
Duct Entry Loss (8 x 19) 
Other Losses 

Hood Static Pressure (16 + 20 + 21) “wg 


Straight Duct Length feet 4 34 
Friction Factor (H,) 0.0098 0.0425 0.0478 0.0299 0.0088 


Friction Loss per VP (23 x 24) 
Number of 90 deg. Elbows 

Elbow Loss Coefficient 

Elbow Loss per VP (26 x 27) 
Number of Branch Entries 


a=f/oafa 
DOINED 


= 
© 


H 

re) 

O 

D 

iS) 

u 
u | 
¢ 

| 

1) 

N 


nN 
fo) 


= 
a 


jah i 
lee} 


4 


(es) 
iO 


~~ 
aN 
° 
= 
(2) 
~ 
i 
i 


Entry Loss Coefficient : 

Branch Entry Loss per VP (29 x 30) 

32 | Special Fitting Loss Coefficients 

33 | Duct Loss per VP (25 + 28 +31 +32) 32 
Duct Loss (33 x 8) 91 38 

35 1 Duct Segment Static Pressure Loss (22 + 34) 

36 | Other Losses (VP-VP,, etc. 

Cumulative Static Pressure = : 429 


= 


w 
Oo 


& 


w WlWiN TN FNlwHO|TMO PM | MPM] PM 
DIolopray~analan WEnN| = 


CIN 


Governing Static Pressure 


O | Corrected Velocity 


1 | Corrected Velocity Pressure 


42 | Resultant Velocity Pressure "wg 1.34 


= 
f 


— 
nN 
ES 


FIGURE 5-10. Blast gate method 


5-24 Industrial Ventilation 


Temperature Remarks: 


Elevation 


5-D 
0 
3500 
6.05 


fa-D 
5 
3500 
3.62 


D-¢ 
50 
3500 
7.05 


C-E 
12210 
4000 
23.65 
22 
2.6398 
4625 
1.33 


8-F 
1200 
3500 
7.93 


lec | 


3500 


ro 
0.1364 
3667 
0.84 
aes 
an 


at bly lo N = hy ~ 
© i) 8 
nO 

oO 

[eo] 


3500 
5.12 


Perdient Information 
From Shaper? 5 


Fag 
Qeon = Gas sp 
duct 


~ 


0.1963 
3565 
0.79 


0.0668 
3742 
0.87 _ 


0.2673 
3555 
0.79 


0.2673 
4490 


4.26. Q 


Q 
4 2 
VP.= qi VP:+ @? VP, 


=e Duct Material 


Galvanized 0.0155 | 0.533 
Black iron, Aluminum, 

PVC, Stainless steel PSEA Ree | eee 
ents (fabric covered 


Fan Static Pressure 
FAN SP = SPou - SPia - VPin 


5 1 


psoas | 18 
1.25 eae 


105 | 20 
eee es 
oe 
[5 


5 


0.0543 0.0264 


1.67 


1.25 2.0 1.5 


1.89 


1.89 
14 
0.0354 


1.05 


is 


4 


0.0436 0.0840 0.0361 0.0087 


Branch Entry Loss Coefficients 


1.67 


ie 
us 

a 

Ke) 

a ny 

Ko N 


ie) 
(es) 
IND 


38 


(es) 
IND 


B he fa de 
ico 
ps 
LD) 


=. 
foe) 
— 
foe) 
foo) 


90° Round Elbow Loss Coefficients 
(5 piece) 


— 
foe) 


Wlwlwl NIN Tm NP MsMm;m|n ala afoafo 
jefe le] abs |s ||| |a Pee Eile e}~fofe fab [»]-| 


1.01 


wo IN to I IS fe = fon N 
b be (We) 
my 
> 
= 
(ee) 


be] | FW | WFD | & Jw] © 
AlLO)POLoay NED a) hi) w 


in lo met 
Oo IN fee) 
[o) 

Im] JRO 

~ 

i<a) 


(Tal es. 
a > 
ite) on 


97 


1.52 2.11 79 2.85 1.65 


ho 


i No N Im oO a 
- > 


60° elbow = 2/3 loss 
45° elbow = 1/2 loss 
30° elbow = 1/3 Joss 


-1.52 
-2.14 
GATE 


-2.11 -2.90 
-4.22 


GATE 


-4.49 -2.85 -1.65 
-2.85 


GATE 


-2.97 
-3.11 
GATE 


Adapted from Michigan Industrial 
Ventilation Conference (8/96) 


81 81 1.14 1.26 1.13 1.13 92 42 


FIGURE 5-10. Blast gate method (continued) 


Exhaust System Design Procedure 5-25 


VELOCITY PRESSURE METHOD CALCULATION SHEET 


Problem #2 With Blast Gates Class Date 


10-J 
1225 
3500 


10.1 8.01 


0.3491 
3509 
0.77 


Maximum Slot Area ft 
Slot Area Selected ft 
m 


Slot Velocity fp 


2 
2 
Slot Velocity Pressure wg 


—s 
=a 
= Se 
Poe lt 
Slot Loss Coefficient om 
Wie aac 
ae 
| 


AAOrN 


Acceleration Factor 


-_ 
[o) 
° 
5 
=s 

a 


Slot Loss per VP (13 + 14) 
Slot Static Pressure (12 x 15) 


Duct Entry Loss Coefficient (Fig. 5-13) ra a 
Acceleration Factor (1 or 0) ac se 
N | Duct Entry Loss per VP (17 + 18) A 


Duct Entry Loss (8 x 19) “wg | 1.74 


ae 
Other Loseee, 
| 
pet 


5 


O-ANCH DOOT 


1 


1.25 


Hood Static Pressure (16 + 20 + 21) “wg 
Straight Duct Length feet 
Friction Loss per VP (23 x 24) : 
| 
| 


9 | Number of Branch Entries 


2 
0.0307 
5 
19 


aan» — 
oO }co 


Entry Loss Coefficient 

Duct Loss per VP (25 + 28 + 31 + 32) 
Duct Loss (33 x 8) 

Duct Segment Static Pressure Loss (22 + 34) 4.17 
Other Losses (VP-VP,, etc. "wa een sll 

Cumulative Static Pressure "wg 
Corrected Velocity 


41 Corrected Velocity Pressure 


2 
2 


Q 
1 
-1.17 


IN 
on 
lens 


42 | Resultant Velocity Pressure 4.00 


FIGURE 5-10. Blast gate method (continued) 


5-26 Industrial Ventilation 


Temperature Remarks: 


Elevation 


= 


a 


J-H 
2125 
4500 
9.31 


H-K 
16285 
4500 
25.76 


Pertinent Information | 
From Chapter 5 
¥SP 


Qoor = Quesign aiid 


duct 


0.4418 
4810 
1.44 


3.6870 
4417 
1.22 _ 


ho 
fo>) 


Q, | 
VP,=@! VP;+ GE VP; | 


~ 
= 


Straight Duct Friction Loss 
ye 


Qs 


ajofoa}]a 
opt wih 


[ DuctMateial | 
116 | catvanices 


Black iron, Aluminum, 
PVC, Stainless steel 


Fan Static Pressure 
FAN SP = SP, - SPin - VPin 


0.0259 0.0072 


1.67 1.0 


— 
= 
[<e) 


nD 
IND 
jan 
Ko 


= 


90° Round Elbow Loss Coefficients 
(5 piece) 


ee Ee 
ieee en, 
a 
ee ee 
coe Cane 
at ae 
Pree eae 
Rte pee 
Saas Rae 
as ee 
ae Pe 
ee eae 
ai Rtas 
Fae thee tees 
ae Fae 
Loose | o.o072_| 
j.21 fog 
[167 tao | 
fe Se tele tye 
Freee Eee 
ia ee 
i) eee 
Pitcwee 


BRIRLOLW! WEWD LW W]lWwIlLWEWLlWINTNINMINIT NIN INEM] NM] — >i 
“—1FOPOILOINEDI OTR LOI MLR LOLOLOaINIOD{ Al BR lOEM | —1O corn 


Loss Coefficients 


my 
io 
panty |e 


& tp 
oO +b 
Be 


60° elbow = 2./3 loss 
45° elbow = 1/2 loss 
30° elbow = 1/3 loss 


Adapted from Michigan Industrial 
Ventilation Conference (8/96) 


: b 
.@al 
IND 

(ee) 


1.29 42 


FIGURE 5-10. Blast gate method (continued) 


corrections to the system design by the following means 
should be made. 


The density variation equations of Chapter | (Section 1.4) 
demonstrate that if temperature increases or absolute pressure 
decreases, the density will decrease. For the mass flow rate at 
the hood(s) to remain the same, the flow rate must change if 
density changes. It is helpful to remember that a fan connected 
to a given system will exhaust the same volume flow rate 
regardless of air density. The mass of air moved, however, 
will be a function of the density. 


5.13.1 Variable Temperature and/or Different Altitude: 
Consider an exhaust system at sea level where 5000 cfm 
of air at 70 F is drawn into a hood. The air is then heated to 
600 F and the density of the air leaving the heater becomes 
0.0375 lbm/ft?. The flow rate downstream of the heater would 
be 10,000 actual cubic feet per minute (acfm) at the new 
density of 0.0375 lbm/ft?. This is true because the 50% 
decrease in density must correspond to a twofold increase in 
the volume flow rate since the mass flow rate has remained 
constant. 


If this temperature effect is ignored and a fan selected for 
5000 cfm is placed in the system, the hood flow rate will be 
well below that required to maintain contaminant control. The 
exact operating point of such a system would have to be 
recalculated based upon the operating point of the incorrectly 
sized fan. 


5.13.2 Elevated Moisture: When air temperature is under 
100 F, no correction for humidity is necessary. When air 
temperature exceeds 100 F and moisture content is greater 
than 0.02 Ibs H,O per pound of dry air, correction is required 
to determine fan operating RPM and power. Correction coef- 
ficients may be read from the psychrometric charts such as 
those illustrated in Figures 5-17 through 5-20. 


5.13.3 Psychrometric Principles: The properties of 
moist air are presented on the psychrometric chart at a single 
pressure. These parameters define the physical properties of 
an air/water vapor mixture. The actual gas flow rate and the 
density of the gas stream at the inlet of the fan must be known 
in order to select the fan. The psychrometric chart provides 
the information required to calculate changes in the flow rate 
and density of the gas as it passes through the various exhaust 
system components. These properties are: 


° Dry-Bulb Temperature is the temperature observed 
with an ordinary thermometer. Expressed in degrees 
Fahrenheit, it may be read directly on the chart and is 
indicated on the bottom horizontal scale. 


° Wet-Bulb Temperature is the temperature at which 
liquid or solid water, by evaporating into air, can bring 
the air to saturation adiabatically at the same 
temperature. Expressed in degrees Fahrenheit, it is read 
directly at the intersection of the constant enthalpy line 


Exhaust System Design Procedure 5-27 


with the 100% saturation curve. 


Dew Point Temperature is that temperature at which 
the air in an air/vapor mixture becomes saturated with 
water vapor and any further reduction of dry bulb 
temperature causes the water vapor to condense or 
deposit as drops of water. Expressed in degrees 
Fahrenheit, it is read directly at the intersection of the 
saturation curve with a horizontal line representing 
constant moisture content. 


Percent Saturation curves reflect the mass of moisture 
actually in the air as a percentage of the total amount 
possible at the various dry bulb and moisture content 
combinations. Expressed in percent, it may be read 
directly from the curved lines on the chart. 


Density Factor is a dimensionless quantity which 
expresses the ratio of the actual density of the mixture 
to the density of standard air (0.075 Ibm/ft*). The lines 
representing density factor typically do not appear on 
low-temperature psychrometric charts when relative 
humidity or percent saturation curves are presented. A 
method of calculating the density of the gas defined by 
a point on the chart (when density factor curves are not 
presented) is discussed in Section 5.13.4. 


Moisture Content, or weight of water vapor, is the 
amount of water that has been evaporated into the air. 
In ordinary air, it is very low pressure steam and has 
been evaporated into the air at a temperature 
corresponding to the boiling point of water at that low 
pressure. Moisture content is expressed in grains of 
water vapor per pound of dry air (7000 grains = one 
pound) or pounds of water vapor per pound of dry air 
and is read directly from a vertical axis. 


Enthalpy (Total Heat) as shown on the psychrometric 
chart is the sum of the heat required to raise the 
temperature of a pound of air from 0 F to the dry-bulb 
temperature, plus the heat required to raise the 
temperature of the water contained in that pound of air 
from 32 F to the dew point temperature, plus the latent 
heat of vaporization, plus the heat required to superheat 
the vapor in a pound of air from the dew point 
temperature to the dry-bulb temperature. Expressed in 
British Thermal Units per pound of dry air, it is shown 
by following the diagonal wet-bulb temperature lines. 


Humid Volume is the volume occupied by the air/vapor 
mixture per pound of dry air and is expressed in cubic 
feet of mixture per pound of dry air. It is most important 
to understand the dimensions of this parameter and 
realize that the reciprocal of humid volume is not 
density. Humid volume is the parameter used most 
frequently in determining flow rate changes within a 
system as a result of mixing gases of different properties 


5-28 Industrial Ventilation 


Dryer 


FIGURE 5-11. System layout 


or when evaporative cooling occurs within the system. 


5.13.4 Density Determination: When the quality of an 
air/vapor mixture is determined by a point on a psychrometric 
chart having a family of density factor curves, all that must 
be done to determine the actual density of the gas at the 
pressure reference for which the chart is drawn is to multiply 
the density factor taken from the chart by the density of 
standard air (0.075 Ibm/ft?). Should relative humidity curves 
be presented on the chart in lieu of density factor curves, 
information available through dimensional analysis must be 
used to determine the actual density of the mixture. This can 
be done quite easily as follows: The summation of one pound 
of dry air plus the mass of the moisture contained within that 
pound of dry air divided by the humid volume will result in 
the actual density of the mixture. 


ee 1+W a 
HV [ = ] 
where: 
p= density of the mix (lbm/ft*) 
W = moisture content (Ibm H,O/lbm dry air) 
HV = humid volume (ft? mix/Ibm dry air) 
5.13.5 Hood Flow Rate Changes with Density: If the 


density of the air entering a hood is different from standard 
density due to changes in elevation, ambient pressure, tem- 
perature, or moisture, the flow rate through the hood should 
be changed to keep the mass flow rate the same as for standard 
air. This can be accomplished by multiplying the hood flow 
rate required for standard air by the ratio of the density of 
standard air to the actual ambient density. 


The example shown in Figure 5-11 illustrates the effect of 
elevated moisture and temperature and a method of calculation: 
EXAMPLE 


GIVEN: The exit flow rate from a 60" x 24' dryer is 16,000 
scfm plus removed moisture. The exhaust air temperature is 
500 F. The drier delivers 60 tons/hr of dried material with 


— ¢(_ R=1.5D (4 piece) | 
t 


\ 


Wet collector 


capacity to remove 5% moisture. Required suction at the dryer 
hood is — 2.0 "wg; minimum conveying velocity must be 4000 
fpm (see Figure 5-11). 


It has been determined that the air pollution control system 
should include a cyclone for dry product recovery and a 
high-energy wet collector. These devices have the following 
operating characteristics: 


¢ Cyclone: Pressure loss is 4.5 "wg at rated flow rate of 
35,000 scfm. The pressure loss across any cyclone 
varies directly with any change in density and as the 
square of any change in flow rate from the rated 
conditions. 


High-Energy Wet Scrubber: The manufacturer has 
determined that a pressure loss of 20 "wg is required in 
order to meet existing air pollution regulations and has 
sized the collector accordingly. The humidifying 
efficiency of the wet collector is 90%. 


NOTE: As a practical matter, a high energy scrubber 
as described in this example would have essentially 
100% humidifying efficiency. The assumption of 90% 
humidifying efficiency along with a high pressure drop 
allows discussion of multiple design considerations in 
one example and was therefore adopted for instruc- 
tional purposes. 


° Fan: A size#34 "XYZ" fan with the performance shown 
in Table 5-3 has been recommended. 


REQUIRED: 

Size the duct and select fan RPM and motor size. 
SOLUTION: 
Step 1 


Find the actual gas flow rate that must be exhausted from 
the dryer. This flow rate must include both the air used for 
drying and the water, as vapor, which has been removed from 


the product. Since it is actual flow rate, it must be corrected 
from standard air conditions to reflect the actual moisture, 
temperature, and pressures which exist in the duct. 


Step 1A: 
Find the amount (weight) of water vapor exhausted. 


Dryer discharge = 60 tons/hr of dried material (given) 


Since the dryer has capacity to remove 5% moisture, the 
dryer discharge is 95% x dryer feed rate. 


60 tons/hr dried material = (0.95) (dryer feed) 


60 tons/hr 


dryer feed = = 63.2 tons/hr 


Moisture removed = (feed rate) — (discharge rate) 


= 63.2 tons/hr — 60 tons/hr 
= 6400 Ibs/hr or 106.7 Ibm/min 


Step 1B 


Find the amount (weight) of dry air exhausted. 


Dry air exhausted = 16,000 scfm at 70 F and 
29.92 “Hg (0.075 Ibsift® density) 


= (16,000 scfm)(0.075 Ibs/ft’) 
= 1200 Ibs/min dry air 


Exhaust rate, lbs/min 


Step 1C 


Knowing the water-to-dry air ratio and the temperature of 
the mixture, it is possible to determine other quantities of the 
air-to-water mixture. This can be accomplished by the use of 
psychrometric charts (see Figures 5-17 to 5-20) which are 
most useful tools when working with humid air. 


W = 0.089 Ibs H2O/Ib dry air 
Dry bulb temperature = 500 F (given) 


Exhaust System Design Procedure 5-29 


The intersection of the 500 F dry-bulb temperature line and 
the 0.089 Ibs H,O/lb dry air line can be located on the 
psychrometric chart (see Figure 5-12). Point #1 completely 
defines the quality of the air and water mixture. Other data 
relative to this specific mixture can be read as follows: 


Dew Point Temperature: 122 F 

Wet-Bulb Temperature: 145 F 

Humid volume, ft? of mixilb of dry air: 27.5 ft°/lb dry air 
Enthalpy, BTU/b of dry air: 235 BTU/Ib dry air 
Density factor, df: 0.53 


Step 1D 


Find actual gas flow rate, (acfm). 


Exhaust flow rate, acfm = (humid volume)(weight of dry 
air/min). Humid volume, HV, was found in Step IC as 27.5 
f8/lb. Weight of dry air/min was found in Step 1B as 1200 
lb/min. Exhaust flow rate = (27.5 fi?/Ib)(1200 Ib/min) = 
33,000 acfm. 


Step 2 


Size the duct. Minimum conveying velocity of 4000 fpm 
was given. Suction at the dryer exit of -2.0 “wg corresponds 
to hood suction. 


The duct area equals the actual flow rate divided by the 
minimum duct velocity, or A = 33,000 + 4,000 = 8.25 ft?. A 
38" diameter duct with a cross-sectional area of 7.876 ft 
should be chosen as this is the largest size available with an 
area smaller than calculated. Then the actual duct velocity 
would be 33,000 actual ft?/min + 7.876 ft? = 4,190 fpm. 


Step 2A 


The velocity pressure in the duct cannot be found using the 
equation VP = (V + 4005)’, as this equation is for standard air 


— 
TABLE 5-3. Fan Rating Table 


Fan size No. 34 Inlet diameter = 34" Max. safe rpm = 1700 
oer 


oases 
20" SP 22" SP 


36" SP [38° SP 40" SP 


fa 
CFM | RPM] BHP | RPM] BHP 


14688 | 1171 | 73.3 | 1225 | 81.4 
16524 | 1181 | 81.8 | 1234 | 90.2 
18360 | 1191 | 90.2 | 1244 sul 


RPM | BHP | RPM| BHP hic Lesa 


1552! 143 | 1594) 153 [1634 162 
1587] 155 | 1600} 165 | 1639} 175 
1565} 167 | 1606} 178 } 1645} 188 


20196 | 1204 | 99.9 | 1256} 109 
22032 11217) 110 | 1268) 120 
23868 | 1230| 120 |1282| 131 


1574} 181 ets 191 | 1654} 202 
1584] 196 | 1624) 207 | 1663| 218 
1594] 211 | 1633} 223 | 1672) 235 


25704 | 1245 131 | 1296] 143 
27540 |1261| 143 [1341 | 156 
29376 | 1277 156 | 13271} 169 


1606) 227 | 1645| 239 1683] 252 
1618) 245 | 1658) 258 | 1695) 271 
1631} 263 |1670| 277 | | 


31212 (1295) 170 | 1344} 184 
33048 | 1313] 184 | 1361) 198 
34884 | 1331 


al 282 |1683 297 
1659} 302 | 1697 | 317 
1674} 323 


5-30 Industrial Ventilation 


“0 ~/ : j ? i 
e ~Humid Yolurne ~ fi" /lb Ory Air 
Ses Pontz, | 
6 < S Ze: me. / i o \ Density Factor~Mixture 
[a a aa 
|< . 
See cl rm 
S ae se \ ra) 
a az Se ee 
ies Or she 
w + fy 
(ae fol 
oO) H 
an se | E T 4 
30 — Sol y—~ POINT 1. 
35% \ ta 
I 
r a2) BS ees eae oO SS ay 1 “SAK 
20 - i 5 \ ey ~\ | \ < So 
ea \a \ 8 U 
i Sa \ oe 
|e | 20, 
Vy l/7, 


20002~C~*” 


500 


Dry Bulb Temperature, F 


FIGURE 5-12. Psychrometric chart for humid air (see Figures 5-17 through 5-20) 


only. The actual velocity pressure in the duct is given by 


VPactuat = (df)(VPsta) 


where: df = density factor 


As the density factor was determined in Step 1C, the actual 
velocity pressure in the duct will be 


VP = (df)(VPsta) = (0.53)(1.09 "wg) = 0.58 "wg 
Step 3 


Calculate the pressure loss from A to B and determine static 
pressure at Point B. 

The data from Figure 5-14 and Table 5-5 can be used 
directly. The static pressure loss through the duct can be found 
by multiplying the length of duct by the friction coefficient, 
adding the elbow loss coefficient, and multiplying the result 
by the duct velocity pressure: 

SP loss = [(0.0045)(30) + 0.27][0.58] = (0.405)(0.58) 

= 0.23 "wg 


Then the static pressure at the inlet to the cyclone should 
be —2.23 "wg (hood suction plus friction and fitting losses). 


Step 4 


The pressure loss of the cyclone is provided by the manu- 
facturer. In this example, the cyclone pressure loss is 4.5 "wg 
at a rated flow of 35,000 scfm. The pressure loss through a 
cyclone, as with duct, varies as the square of the change in 
flow rate and directly with change in density. 


Therefore, the actual loss through the cyclone would be 


33,000 
35,000 


2 
(45)| (0.53) = -212 “wg 


and the static pressure at the cyclone outlet would be ~4.35 
"we. 
Step 5 

The calculation from Point C to D is the same as from A to 


B in Step 3. Thus, the static pressure at the wet collector inlet 
would be 


4.35 — (0.0045)(15)(0.58) = 4.39 "wg 


NOTE: Information for Steps 6 and 7 which involve 
calculation of changes in flow rate, density, etc., across 
the wet collector should be provided by the equipment 
manufacturer. 


Step 6 


An important characteristic of wet collectors is their ability 
to humidify a gas stream. The humidification process is 
generally assumed to be adiabatic (without gain or loss of heat 
to the surroundings). Therefore, water vapor is added to the 
mixture, but the enthalpy, expressed in BTU/Ib dry air, re- 
mains unchanged. During the process of humidification, the 
point on the psychrometric chart that defines the quality of 
the mixture moves to the left, along a line of constant enthalpy, 
toward saturation. 


All wet collectors do not have the same ability to humidify. 
If a collector is capable of taking an air stream to complete 
adiabatic saturation, it is said to have a humidifying efficiency 
of 100%. The humidifying efficiency of a given device may 
be expressed by either of the following equations: 
wets x 100 


Tn 


i s 


where: 


Nn = humidifying efficiency, % 
T 
T, = dry-bulb temperature at collector outlet, F 


dry-bulb temperature at collector inlet, F 


T, = adiabatic saturation temperature, F 


or 
= Wee x 100 
i W, 

where: 
W, = moisture content in Ib H,O/Ib dry air at inlet 
W, = moisture content in Ib H,O/lb dry air at outlet 
W, = moisture content in lb H,O/Ib dry air at adiabatic 

saturation conditions 
Step 6A 


Find the quality of the air to water mixture at Point 2, the 
collector outlet. 


Humidifying Efficiency = 90% (given). Dry-bulb Tem- 
perature at Collector Inlet = 500 F (given). Adiabatic satura- 
tion temperature = 145 F from inspection of Psychrometric 
Chart. 

_ {500 - to) 


90 = "2 x 100 
(500 — 145) 


where: 
to = 180 F 

Then the air leaving the collector will have a dry-bulb tem- 
perature of 180 F and an enthalpy of 235 BTU/Ib of dry air as 
the humidifying process does not change the total heat or 
enthalpy. 

The point of intersection of 180 F dry bulb and 235 BTU/Ib 
dry air on the psychrometric chart defines the quality of the 


air leaving the collector and allows other data to be read from 
the chart as follows: 


Dew Point Temperature 143 F 

Wet-Bulb Temperature 145 F 

Humid Volume, ft?/lb dry air 20.5 ft°b dry air 
Enthalpy, BTU/Ib dry air 235 BTU/Ib dry air 
Density factor, df 0.76 


Step 7 


What is the exhaust flow rate in acfm and the density factor 
at the collector outlet? 


Step 7A 


Exhaust flow rate = (humid volume)(weight. of dry 
air/min). Humid Volume from Step 6 is 20.5 ft?/Ib dry air. 
Weight of dry air/min from Step 1B is 1200 Ibs/min. Flow 
rate = (20.5 ft?/Ib)(1200 lbs/min) = 24,600 acfm. 


Exhaust System Design Procedure 5-31 


As the wet collector loss was stated to be 20 “wg, the static 
pressure at the wet collector outlet would be —24.39 ”weg. 


Step 7B 


On low-pressure exhaust systems, where the negative pres- 
sure at the fan inlet is less than 20 “we, the effect of the 
negative pressure is usually ignored. However, as the pres- 
sures decrease, or the magnitude of negative pressures in- 
creases, it is understood that gases expand to occupy a larger 
volume. Unless this larger volume is anticipated and the fan 
sized to handle the larger flow rate, it will have the effect of 
reducing the amount of air that is pulled into the hood at the 
beginning of the system. From the characteristic equation for 
the ideal gas laws, PQ = wRT (where w = the mass flow rate 
in !bm/min), the pressure flow rate relationship is 


P4Q1 = P2Q2 
or 

Pr Qe 

P,Q, 


Up to this point, the air has been considered to be at standard 
atmospheric pressure which is 14.7 psia, 29.92 “Hg or 407 
weg. The pressure within the duct at Point E is —24.4 “wg and 
minus or negative only in relation to the pressure outside the 
duct which is 407 ’wg. Therefore, the absolute pressure 
within the duct is 407 “wg -24.4 "wg = 382.6 “wg. 


407 Q, 


382.6 24.600 cfm 
Q2 = 26,170 acfm 
Step 7C 


Pressure also affects the density of the air. From PQ =wRT 
the relationship 


can be derived. Density factor is directly proportional to the 
density and the equation can be rewritten 


Substitute 


407 0.76 
3826 df, 


(df, was determined to be 0.76 in Step 6.) 
dfo= 0.71 
Step 7D 


The duct from the wet collector to the fan can now be sized. 


§-32 industrial Ventilation 


The flow rate leaving the wet collector was 26,170 acfm. As 
the fan selected has a 34-in. diameter inlet (area = 6.305 ft?), 
it is logical to make the duct from the wet collector to the fan 
a 34-in. diameter. Thus, the velocity through the duct would 
be 26,170 + 6.305 = 4,151 fpm. The VP would be (0.71)(4151 
+ 4005)? = (0.71)(1.07) = 0.76 "weg. 


Step 7E 


The duct pressure loss, based on 26,170 cfm and a 34-in. 
diameter duct, would be (0.0052)(5)(0.76) = 0.02 "wg. There- 
fore, the SP at the fan inlet would be —24.41 "wg, 


Step 8 


Calculate the pressure loss from fan discharge F to stack 
discharge G. Since the air is now on the discharge side of the 
fan, the pressure is very near atmospheric. No pressure cor- 
rection is needed. The flow rate and density factor are 24,600 
acfm and 0.76, respectively. 


Assuming that the fan discharge area is nearly the same as 
at the fan inlet, the same 34-in. diameter duct would result in 
a velocity of 3902 fpm. The velocity pressure would be 
(0.76)(3902 + 4005)? = 0.72 "we. 


From Table 5-5, the friction coefficient is 0.0052 and the 
frictional pressure loss for the 30 ft. high stack would be 
(0.0052)(30)(0.72) = 0.11 "wg. As the static pressure at the 
exit of the stack must be atmospheric, the static pressure at 
the fan exit will be positive. 


Step 9 
Determine actual fan static pressure. 


Actual FSP = SPout - SPin — VP in 
= + 0.11 -(-24.41) - 0.76 
= 23.76 "wg 


Step 10 


Determine equivalent fan static pressure in order to enter 
fan rating table. Equivalent fan static pressure is determined 
by dividing the actual fan static pressure by the density factor 
at the fan inlet. This is necessary since fan rating tables are 
based on standard air. 


Equivalent FSP = on = 33.46 "wg 


Step 11 


Select fan from rating table using the equivalent fan SP and 
the fan inlet flow rate. Interpolating the fan rating table (Table 
5-3) for 26,200 cfm at 33.5 "wg yields a fan speed of 1559 
RPM at 217 BHP. 


Step 12 


Determine the actual required fan power. Since actual 


density is less than standard air density, the actual required 
power is determined by multiplying by the density factor, or 
(217 BHP)(0.71) = 154 BHP. If a damper is installed in the 
duct to prevent overloading of the motor, at cold start the 
motor need only be a 200 HP (see Chapter 6). 


5.14 AIR CLEANING EQUIPMENT 


Dusts, fumes, and toxic or corrosive gases should not be 
discharged to the atmosphere. Each exhaust system handling 
such materials should be provided with an adequate air cleaner 
as outlined in Chapter 4. As a rule, the exhaust fan should be 
located on the clean air side of such equipment. An exception 
is in the use of cyclone cleaners where the hopper discharge 
is not tightly sealed and better performance is obtained by 
putting the fan ahead of the collector. 


5.15 EVASE DISCHARGE 


An evasé discharge is a gradual enlargement at the outlet 
of the exhaust system (see Figure 5-16). The purpose of the 
evasé is to reduce the air discharge velocity efficiently; thus, 
the available velocity pressure can be regained and credited 
to the exhaust system instead of being wasted. Practical 
considerations usually limit the construction of an evasé to 
approximately a 10° angle (5° side angle) and a discharge 
velocity of about 2000 fpm (0.25 "wg velocity pressure) for 
normal exhaust systems. Further streamlining or lengthening 
the evasé yields diminishing returns. 


It should be noted, however, that for optimum vertical 
dispersion of contaminated air, many designers feel that the 
discharge velocity from the stack should not be less than 
3000-3500 fpm. When these considerations prevail, the use 
of an evasé is questionable. 


The following example indicates the application of the 
evasé fitting. It is not necessary to locate the evasé directly 
after the outlet of the fan. It should be noted that, depending 
upon the evasé location, the static pressure at the fan discharge 
may be below atmospheric pressure, i.e., negative (—), as 
shown in this example. 


EXAMPLE 
Duct No. Dia. Q V VP SP 
1 Fan inlet 20 8300 3800 090 -7.27 
2 Fan Discharge = 8300 3715 0.86 

16.5 x 19.5 
3 Round Duct Connection 20 3800 0.90 


4 _Evasé Outlet 28 1940 0.23 0 


To calculate the effect of the evasé, see Figure 5-16 for 
expansion at the end of the duct where the Diameter Ratio, 
D,+D3 = 28+20 = 1.4 and Taper length L/D = 40+20 = 2. 


R = 0.52 x 70% (since the evasé is within 5 
diameters of the fan outlet) 


VP3 =0.9 as given 


SP4 =0 (since the end of the duct is at atmospheric 
pressure) 


SP3 = SP4—R(VP3) 
= 0 — (0.52)(0.70)(0.90") 
= -0.33 "wg 


FSP = SPoutiet — SPintet — VPinlet 
= -0.33 — (-7.27) — 0.9 = 6.04 "wg 


5.16 EXHAUST STACK OUTLETS 


The final component of the ventilation system is the ex- 
haust stack, an extension of the exhaust duct above the roof. 
There are two reasons for the placement of an exhaust stack 
on a ventilation system. First, the air exhausted by a local 
exhaust system should escape the building envelope. Second, 
once it has escaped the building envelope, the stack should 
provide sufficient dispersion so that the plume does not cause 
an unacceptable situation when it reaches the ground. This 
brief description of stack design will address only the first 
concern. 


When placing an exhaust stack on the roof of a building, 
the designer must consider several factors. The most impor- 
tant is the pattern of the air as it passes the building. Even in 
the case of a simple building design with a perpendicular 
wind, the air flow patterns over the building can be complex 
to analyze. Figure 5-28a shows the complex interaction be- 
tween the building and the wind at height H. A stagnation 
zone is formed on the upwind wall. Air flows away from the 
stagnation zone resulting in a down draft near the ground. 
Vortices are formed by the wind action resulting in a recircu- 
lation zone along the front of the roof or roof obstructions, 
down flow along the downwind side, and forward flow along 
the upwind side of the building. 


Figure 5-28b shows a schematic of the critical zones 
formed within the building cavity. A recirculation zone is 
formed at the leading edge of the building. A recirculation 
zone is an area where a relatively fixed amount of air moves 
in a circular fashion with little air movement through the 
boundary. A stack discharging into the recirculation zone can 
contaminate the zone. Consequently, all stacks should pene- 
trate the recirculation zone boundary. 


The high turbulence region is one through which the air 
passes; however, the flow is highly erratic with significant 
downward flow. A stack that discharges into this region will 
contaminate anything downwind of the stack. Consequently, 
all stacks should extend high enough that the resulting plume 
does not enter the high turbulence region upwind of an air 
intake. 


Because of the complex flow patterns around simple build- 
ings, it is almost impossible to locate a stack that is not 
influenced by vortices formed by the wind. Tall stacks are 
often used to reduce the influence of the turbulent flow, to 


Exhaust System Design Procedure 5-33 


release the exhaust air above the influence of the building and 
to prevent contamination of the air intakes. Selection of the 
proper location is made more difficult when the facility has 
several supply and exhaust systems and when adjacent build- 
ings or terrain cause turbulence around the facility itself. 


When locating the stack and outdoor air inlets for the air 
handling systems, it is often desirable to locate the intakes 
upwind of the source. However, often there is no true upwind 
position. The wind in all locations is variable. Even when 
there is a natural prevailing wind, the direction and speed are 
constantly changing. If stack design and location rely on the 
direction of the wind, the system will clearly fail. 


The effect of wind on stack height varies with speed: 


e Atvery low wind speeds, the exhaust jet from a vertical 
stack will rise above the roof level resulting in 
significant dilution at the air intakes. 


° Increasing wind speed will decrease plume rise and 
consequently decrease dilution. 


e Increasing wind speed will increase turbulence and 
consequently increase dilution. 


The prediction of the location and form of the recirculation 
cavity, high turbulence region and roof wake is difficult. 
However, for wind perpendicular to a rectangular building, 
the height (H) and the width (W) of the upwind building face 
determine the airflow patterns. The critical dimensions are 
shown in Figure 5-28b. According to Wilson,©® the critical 
dimensions depend on a scaling coefficient (R) which is given 
by: 


R= Bos? x Bes [5.9] 


where B, is the smaller and B, is the larger of the dimensions 
H and W. When B, is larger than 8B,, use B, = 8 B, to calculate 
the scaling coefficient. For a building with a flat roof, Wil- 
son®” estimated the maximum height (H,), center (X,), and 
lengths (L,) of the recirculation region as follows: 


Hc = 0.22R [5.10] 
Xc=0.5R [5.11] 
lc=0.9R [5.12] 


In addition, Wilson estimated the length of the building 
wake recirculation region by: 


L-=1.0R [5.13] 


The exhaust air from a stack often has not only an upward 
momentum due to the exit velocity of the exhaust air but 
buoyancy due to its density as well. For the evaluation of the 
stack height, the effective height is used (see Figure 5-29a). 
The effective height is the sum of the actual stack height (H,), 
the rise due to the vertical momentum of the air, and any wake 
downwash effect that may exist. A wake downwash occurs 
when air passing a stack forms a downwind vortex. The vortex 


5-34 Industrial Ventilation 


will draw the plume down, reducing the effective stack height 
(see Figure 5-29b). This vortex effect is eliminated when the 
exit velocity is greater than 1.5 times the wind velocity. If the 
exit velocity exceeds 3000 fpm, the momentum of the exhaust 
air reduces the potential downwash effect. 


The ideal design extends the stack high enough that the 
expanding plume does not meet the wake region boundary. 
More realistically, the stack is extended so that the expanding 
plume does not intersect the high turbulence region or any 
recirculation cavity. According to Wilson, ©® the high turbu- 
Jence region boundary (Z3) follows a 1:10 downward slope 
from the top of the recirculation cavity. 


To avoid entrainment of exhaust gas into the wake, stacks 
must terminate above the recirculation cavity. The effective 
stack height to avoid excessive reentry can be calculated by 
assuming that the exhaust plume spreads from the effective 
stack height with a slope of 1:5 (see Figure 5-28b). The first 
step is to raise the effective stack height until the lower edge 
of the 1:5 sloping plume avoids contact with all recirculation 
zone boundaries. The zones can be generated by rooftop 
obstacles such as air handling units, penthouses or architec- 
tural screens. The heights of the cavities are determined by 
Equations 5.10, 5.11 and 5.12 using the scaling coefficient for 
the obstacle. Equation 5.13 can be used to determine the 
length of the wake recirculation zone downwind of the obsta- 
cle. 


If the air intakes, including windows and other openings, 
are located on the downwind wall, the lower edge of the plume 
with a downward slope of 1:5 should not intersect with the 
recirculation cavity downwind of the building. The length of 
the recirculation cavity (L,) is given by Equation 5.13. If the 
air intakes are on the roof, the downward plume should not 
intersect the high turbulence region above the air intakes. 
When the intake is above the high turbulence boundary, 
extend a line from the top of the intake to the stack with a 
slope of 1:5. When the intake is below the high turbulence 
region boundary, extend a vertical line to the boundary, then 
extend back to the stack with a slope of 1:5. This allows the 
calculation of the necessary stack height. The minimum stack 
height can be determined for each air intake. The maximum 
of these heights would be the required stack height. 


In large buildings with many air intakes, the above proce- 
dure will result in very tall stacks. An alternate approach is to 
estimate the amount of dilution that is afforded by stack 
height, distance between the stack and the air intake and 
internal dilution that occurs within the system itself. This 
approach is presented in the "Airflow Around Buildings" 
chapter in the Fundamentals volume of the 1993 ASHRAE 
Handbook.©® 


5.16.1 Stack Considerations: 


1. Discharge velocity and gas temperature influence the 
effective stack height. 


. Wind can cause a downwash into the wake of the stack 


reducing the effective stack height. Stack velocity 
should be at least 1.5 times the wind velocity to prevent 
downwash. 


. A good stack velocity is 3000 fpm because it: 


¢ Prevents downwash for winds up to 2000 fpm (22 
mph). Higher wind speeds have significant dilution 
effects. 


° Increases effective stack height. 


° Allows selection of a smaller centrifugal exhaust 
fan to provide a more stable operation point on the 
fan curve (see Chapter 6). 


° Provides conveying velocity if there is dust in the 
exhaust or there is a failure of the air cleaning 
device. 


. High exit velocity is a poor substitute for stack height. 


For example, a flush stack requires a velocity over 
8000 fpm to penetrate the recirculation cavity bound- 
ary. 


. The terminal velocity of rain is about 2000 fpm. A 


stack velocity above 2600 fpm will prevent rain from 
entering the stack when the fan is operating. 


. Locate stacks on the highest roof of the building when 


possible. If not possible, a much higher stack is re- 
quired to extend beyond the wake of the high bay, 
penthouse, or other obstacle. 


. The use of an architectural screen should be avoided. 


The screen becomes an obstacle and the stack must be 
raised to avoid the wake effect of the screen. 


. The best stack shape is a straight cylinder. If a drain is 


required, a vertical stack head is preferred (see Figure 
5-30). In addition, the fan should be provided with a 
drain hole and the duct should be slightly sloped 
toward the fan. 


. Rain caps should not be used. The rain cap directs the 


air toward the roof, increases the possibility of reentry, 
and causes exposures to maintenance personnel on the 
roof. Moreover, rain caps are not effective. A field 
study©” with a properly installed standard rain cap 
showed poor performance. A 12-inch diameter stack 
passed 16% of all rain and as high as 45% during 
individual storms. 


. Separating the exhaust points from the air intakes can 


reduce the effect of reentry by increasing dilution. 


. In some circumstances, several small exhaust systems 


can be manifolded to a single exhaust duct to provide 
internal dilution thereby reducing reentry. 


. A combined approach of vertical discharge, stack 


height, remote air intakes, proper air cleaning device, 
and internal dilution can be effective in reducing the 


consequences of reentry. 


13. A tall stack is not an adequate substitute for good 
emission control. The reduction achieved by properly 
designed air cleaning devices can have a significant 
impact on the potential for reentry. 


§.17 AIR BLEED-INS 


Bleed-ins are used at the ends of branch ducts to provide 
additional air flow rates to transport heavy material loads as 
in woodworking at saws and jointers or at the ends of a main 
duct to maintain minimum transport velocity when the system 
has been oversized deliberately to provide for future expan- 
sion. Some designers use bleed-ins also to introduce addi- 
tional air to an exhaust system to reduce air temperature and 
to assist in balancing the system. 


EE 


EXAMPLE 


End cap bleed-in (see sketch). Consider it to be an orifice 
or slot. From Figure 5-13, h, = 1.78 VP. 


1. Calculate SP for branch duct to junction (X). 


2. Determine flow rate in main duct according to design 
or future capacity or determine Q bleed-in directly 
from temperature or moisture considerations. 


3. Q bleed-in = (Q main duct) — (Q branch) 


4. SP bleed-in = SP branch as calculated = (h, + 1 VP) = 
(1.78 + 1.0) VP 
5. VP, bleed-in = —SP__ = SP 


(178+10) 278 
6. Velocity, bleed-in from VP and Table 5-7a. 


: Q bleed -in 
7. Area bleed-in = Viblecd cin 


5.18 OPTIMUM ECONOMIC VELOCITY 


In systems which are intended to carry dust, a minimum 
conveying velocity is necessary to ensure that the dust will 
not settle in the duct. Also, when a system is installed in a 
quiet area, it may be necessary to keep velocities below some 
maximum to avoid excessive duct noise. When axial flow fans 
are used, duct velocities of 1000 to 1500 fpm are preferred. 
In a gas or vapor exhaust system installed in a typical factory 


Exhaust System Design Procedure 5-35 


environment where none of these restrictions apply, the ve- 
locity may be selected to yield the lowest annual operating 
cost. 


To determine the optimum economic velocity, the system 
must first be designed at any assumed velocity and the total 
initial costs of duct material, fabrication, and installation 
estimated.6') 


This optimum economic velocity may range from under 
2000 fpm to over 4000 fpm. Lengthy expected service periods 
and system operating times tend to lower the optimum while 
high interest rates and duct costs tend to raise the optimum. 
In general, a velocity of 2500 to 3000 fpm will not result in 
equivalent total annual costs much in excess of the true 
optimum. 


5.19 CONSTRUCTION GUIDELINES FOR LOCAL 
EXHAUST SYSTEMS 


Ducts are specified most often for use in the low static 
pressure range (-10 "wg to +10 "wg), but higher static pres- 
sures are occasionally encountered. The duct conveys air or 
gas which is sometimes at high temperatures and often con- 
taminated with abrasive particulate or corrosive aerosols. 
Whether conditions are mild or severe, correct design and 
competent installation of ducts and hoods are necessary for 
proper functioning of any ventilation system. The following 
minimum specifications are recommended. 


Exhaust systems should be constructed with materials suit- 
able for the conditions of service and installed in a permanent 
and workman-like manner. To minimize friction loss and 
turbulence, the interior of all ducts should be smooth and free 
from obstructions — especially at joints. 


5.19.1 Materials: Ducts are constructed of black iron, 
which has been welded, flanged, and gasketed; or of welded 
galvanized sheet steel unless the presence of corrosive gases, 
vapors, and mists or other conditions make such material 
impractical. Arc welding of black iron lighter than 18 gauge 
is not recommended. Galvanized construction is not recom- 
mended for temperatures exceeding 400 F. The presence of 
corrosive gases, vapor, and mist may require the selection of 
corrosive resistant metals, plastics, or coatings. It is recom- 
mended that a specialist be consulted for the selection of 
materials best suited for applications when corrosive atmos- 
pheres are anticipated. Table 5-4 provides a guide for selec- 
tion of materials for corrosive conditions. 


5.19.2 Construction: 


1. There are four classifications for exhaust systems on 
noncorrosive applications: 


Class I — Light Duty: Includes nonabrasive applica- 
tions (e.g., replacement air, general ventilation, gase- 
ous emissions control). 


Class 2 — Medium Duty: Includes applications with 


5-36 


Industrial Ventilation 


moderately abrasive particulate in light concentrations 
(e.g., buffing and polishing, woodworking, grain 
dust). 


Class 3 —- Heavy Duty: Includes applications with 
high abrasive in low concentrations (e.g., abrasive 
cleaning operations, dryers and kilns, boiler breeching, 


sand handling). 


Class 4 — Extra Heavy Duty: Includes applications 
with highly abrasive particles in high concentrations 
(e.g., materials conveying high concentrations of par- 
ticulate in all examples listed under Class 3 — usually 
used in heavy industrial plants such as steel mills, 
foundries, mining, and smelting). 


. For most conditions, round duct is recommended for 


industrial ventilation, air pollution control, and dust 
collecting systems. Compared to non-round duct, it 
provides for lower friction loss, and its higher struc- 
tural integrity allows lighter gauge materials and fewer 
reinforcing members. Round duct should be con- 
structed in accordance with the Reference 5.11. Metal 
thickness required for round industrial duct varies with 
classification, static pressure, reinforcement, and span 
between supports. Metal thicknesses required for the 
four classes are based on design and use experience. 


. Rectangular ducts should only be used when space 


requirements preclude the use of round construction. 
Rectangular ducts should be as nearly square as possi- 
ble to minimize resistance, and they should be con- 
structed in accordance with Reference 5.12. 


. For many applications, spiral wound duct is adequate 


and less expensive than custom construction. How- 
ever, spiral wound duct should not be used for Classes 
3 and 4 because it does not withstand abrasion well. 
Elbows, branch entries, and similar fittings should be 
fabricated, if necessary, to achieve good design. Spe- 
cial considerations concerning use of spiral duct are as 
follows: 


A. Unless flanges are used for joints, the duct should 
be supported close to each joint, usually within 2 
in. Additional supports may be needed. See Refer- 
ence 5.11. 


B. Joints should be sealed by methods shown to be 
adequate for the service. 


C. Systems may be leak tested after installation at the 
maximum expected static pressure. The acceptable 
leakage criteria, often referred to as leakage class, 
should be carefully selected based on the hazards 
associated with the contaminant. 


. The following formula! can be used for specifying 


ducts to be constructed of metals other than steel. For 
a duct of infinite length, the required thickness may be 
determined from: 


t , (1-v*) 
— = 30.035714 p -—— 
D p E (52+D) 


where: 


10. 


t =the thickness of the duct in inches 
D =the diameter of the duct in inches 


p =the intensity of the negative pressure on the 
duct in psi 


E = modulus of elasticity in psi 
v = Poisson's ratio 


The above equation for Class 1 ducts incorporates a 
safety factor which varies linearly with the diameter 
(D), beginning at 4 for small ducts and increasing to 8 
for duct diameters of 60 in. This safety factor has been 
adopted by the sheet metal industry to provide for lack 
of roundness; excesses in negative pressure due to 
particle accumulation in the duct and other manufac- 
turing or assembly imperfections unaccounted for by 
quality control; and tolerances provided by design 
specifications. 


Additional metal thickness must be considered for 
Classes 2, 3 and 4. The designer is urged to consult the 
Sheet Metal and Air Conditioning Contractors Na- 
tional Association (SMACNA) standards for complete 
engineering design procedures. 


. Hoods should be a minimum of two gauges heavier 


than straight sections of connecting branches, free of 
sharp edges or burrs, and reinforced to provide neces- 
sary stiffness. 


. Longitudinal! joints or seams should be welded. All 


welding should conform to the standards established 
by the American Welding Society (AWS) structural 
code.©:'3) Double lock seams are limited to Class 1 
applications. 


. Duct systems subject to wide temperature fluctuations 


should be provided with expansion joints. Flexible 
materials used in the construction of expansion joints 
should be selected with temperature and corrosion 
conditions considered. 


. Elbows and bends should be aminimum of two gauges 


heavier than straight lengths of equal diameter and 
have a centerline radius of at least two and preferably 
two and one-half times the pipe diameter (see Figure 
5-21). Large centerline radius elbows are recom- 
mended where highly abrasive dusts are being con- 
veyed. 


Elbows of 90° should be of a five piece construction 
for round ducts up to 6 in. and of a seven piece 
construction for larger diameters. Bends less than 90° 
should have a proportional number of pieces. Prefab- 
ricated elbows of smooth construction may be used 


Exhaust System Design Procedure 5-37 


TABLE 5-4. Typical Physical and Chemical Properties of Fabricated Plastics and Other Materials 
Resistance to 


Trade Max. Opr.  Flam- Mineral Strong Weak Strong Weak Salt 
Chemical Type Names Temp.,F mability Gasoline Oil Alk. Alk. Acid Acid Solution Solvents 
Urea Formaldehyde _ 170 Self Ext. Good Good Unac. Fair Poor Poor — Good 
jaskon 
Sylplast 
Melamine Silke 210-300 SelfExt. Good Good Poor Good Poor Good — Good 
jaSKON 
Formaldehyde Resknene 
Phenolic Spt 250-450 SelfExt. Fair — Poor Fair Poor Fair ae Fair 
urile 
Durez G.E. 
Resinox 
Alkyd Plaskon _ Self Ext. Good — Unac. Poor — Good — Good 
Silicone Bakelite G.E. 550 _ Good Good — _ Good Good _ Unac. 
Epoxy ae 50-200 SelfExt. Good — Good = Good Good Good _ Good 
raid 
Maraset 
Renite 
Tool Plastik 
Epon Resin 
Cast Phenolic Marblette _ Self Ext. _ — Unac. Fair Good Good — Good to 
Unac. 
Allyl & Polyester nan 300-450 Self Ext. _ _ Poor Fair Poor Fair _ Fair 
AKellte 
Plaskon 
Glykon 
Paraplex 
Acrylic rie 140-200 0.5-2.0 — _ — Good Unac. _—- Good _ Good to 
iene in/min Unac. 
Polyethylene Tenite 140-200 Slow —_ _ — _ _ — — Unac. 
Irrathene Burning 
Tetrafluoroethylene Teflon 500 Non-Fl. Good _ Good Good Good Good _ Good 
Chlorotrifluoroethylene Kel F 
Polyvinyl Formal & anes — Slow Good Good Good Good Unac. — Unac. — Unac. 
ulacite . 
Butyral Saftex Burning 
Butvar 
Formuaré 
Vinyl Chloride Sid Ve 130-175 Slow _— _ Good Good Good Good = Unac. 
Polymer Dow pve Burning 
& Copolymer Vygen 
Vinylidene Chloride Saran 160-200 SelfExt. Good Good Good Good Good Good — Fair 
Styrene sri 150-165 0.5-2.0 Unac. Fair Good Good _ — Good Poor 
Sat in/min 
Dylene 
Luxtrex 
Polystyrene Unac. Fair Good Good _ — Good Poor 
Reinforced with 
Fibrous Glass 
Cellulose Acetate Celanese Thermo 0.5-2.0 Good Good Unac. Unac. Unac. Fair — Poor 
ee Plastic —_in/min 
Tenite 
Nylon af 250 SelfExt. Good Good Good Good  Unac. Good _ Good 
y 
Tynex 
Glass Pyrex 450 Non-Fl. Good Good Good Good Good Good Good Good 


NOTE: Each situation must be thoroughly checked for compatibility of materials during the design phase or if usage is changed. 


5-38 


Industrial Ventilation 


(see Figure 5-22 for heavy duty elbows). 


. Where the air contaminant includes particulate that 


may settle in the ducts, clean-out doors should be 
provided in horizontal runs, near elbows, junctions, 
and vertical runs. The spacing of clean-out doors 
should not exceed 12 ft for ducts of 12 in. diameter and 
less but may be greater for larger duct sizes (see Figure 
5-23). Removable caps should be installed at all ter- 
minal ends, and the last branch connection should not 
be more than 6 in. from the capped end. 


. Transitions in mains and sub-mains should be tapered. 


The taper should be at least five units long for each one 
unit change in diameter or 30° included angle (see 
Figure 5-25). 


. All branches should enter the main at the center of the 


transition at an angle not to exceed 45° with 30° pre- 
ferred. To minimize turbulence and possible particu- 
late fall out, connections should be to the top or side 
of the main with no two branches entering at opposite 
sides (see Figure 5-26). 


. Where condensation may occur, the duct system 


should be liquid tight and provisions should be made 
for proper sloping and drainage. 


. A straight duct section of at least six equivalent duct 


diameters should be used when connecting to a fan (see 
Figure 5-27). Elbows or other fittings at the fan inlet 
will seriously reduce the volume discharge (see Fig- 
ures 6-23, 6-24 and AMCA 2016). The diameter of 
the duct should be approximately equal to the fan inlet 
diameter. 


. Discharge stacks should be vertical and terminate at a 


point where height or air velocity limit re-entry into 
supply air inlets or other plant openings (see Figures 
5-28 and 5-29). 


5.19.3 System Details: 


1. 


Provide duct supports of sufficient capacity to carry 
the weight of the system plus the weight of the duct half 
filled with material and with no load placed on connect- 
ing equipment. [See SMACNA standards.6'!:5:'9)] 


. Provide adequate clearance between ducts and ceil- 


ings, walls and floors for installation and maintenance. 


. Install fire dampers, explosion vents, etc., in accord- 


ance with the National Fire Protection Association 
Codes and other applicable codes and standards. 


. Avoid using blast gates or other dampers. However, if 


blast gates are used for system adjustment, place each 
in a vertical section midway between the hood and the 
junction. To reduce tampering, provide a means of 
locking dampers in place after the adjustments have 
been made. (See Figure 5-24 for types.) 


5. Allow for vibration and expansion. If no other consid- 
erations make it inadvisable, provide a flexible con- 
nection between the duct and the fan. The fan housing 
and drive motor should be mounted on a common base 
of sufficient weight to dampen vibration or on a prop- 
erly designed vibration isolator. 


6. Exhaust fans handling explosive or flammable atmos- 
pheres require special construction (see Section 6.3.9). 


7. Do not allow hoods and duct to be added to an existing 
exhaust system unless specifically provided for in the 
original design or unless the system is modified. 


8. Locate fans and filtration equipment such that mainte- 
nance access is easy. Provide adequate lighting in 
penthouses and mechanical rooms. 


5.19.4 Codes: Where federal, state, or local laws conflict 
with the preceding, the more stringent requirement should be 
followed. Deviation from existing regulations may require 
approval. 


5.19.5 Other Types of Duct Materials: 


1. Avoid use of flexible ducts. Where required, use a 
noncollapsible type that is no longer than necessary. 
Refer to the manufacturer's data for friction and bend 
losses. 


2. Commercially available seamless tubing for small 
duct sizes (i.e., up to 6 in.) may be more economical 
on an installed cost basis than other types. 


3. Plastic pipe may be the best choice for some applica- 
tions (e.g., corrosive conditions at low temperature; 
see Table 5-4.) For higher temperatures, consider fi- 
berglass or a coated duct. 


4. Friction losses for non-fabricated duct will probably 
be different than shown in Tables 5-5 and 5-6. For 
specific information, consult manufacturer's data. 


5.19.6 Testing: The exhaust system should be tested and 
evaluated (see Chapter 9). Openings for sampling should be 
provided in the discharge stack or duct to test for compliance 
with air pollution codes or ordinances. 


REFERENCES 


5.1 Loeffler, J.J.: Simplified Equations for HVAC Duct 
Friction Factors. ASHRAE Journal, pp. 76-79 (Janu- 
ary 1980). 

5.2 Guffey, S.E.: Air-Flow Redistribution in Exhaust 
Ventilation Systems Using Dampers and Static Pres- 
sure Ratios. Appl. Occup. Environ. Hyg. 
8(3):168-177 (March 1993). 


5.3 The Kirk and Blum Manufacturing Co.: Woodwork- 
ing Plants, p. W-9. Kirk and Blum, Cincinnati, OH. 


5.4 Air Movement and Control Association, Inc.: AMCA 


Standard 210-74. AMCA, Arlington Heights, IL. 


5.5 Constance, J.A.: Estimating Air Friction in Triangular 
Ducts. Air Conditioning, Heating and Ventilating 
60(6)85—86 (June 1963). 


5.6 Wilson, D.J.: Contamination of Air Intakes from Roof 
Exhaust Vents. ASHRAE Transactions 82:1024—38 
ASHRAE, Atlanta, GA (1976). 


5.7 Wilson, D.J.: Flow Patterns Over Flat Roof Buildings 
and Application to Exhaust Stack Design. ASHRAE 
Transactions 85:284—-95. ASHRAE, Atlanta, GA 
(1979). 


5.8 American Society of Heating, Refrigerating and Air- 
Conditioning Engineers: 1993 ASHRAE Handbook, 
Fundamentals Volume 14:1-14,18. ASHRAE, At- 
lanta, GA (1993). 


5.9 Clark, John: The Design and Location of Building 
Inlets and Outlets to Minimize Wind Effect and Build- 
ing Reentry. Amer. Indus. Hyg. Assoc. J. 26:262 
(1956). 

5.10 Lynch, J.R.: Computer Design of Industrial Exhaust 
Systems. Heating, Piping and Air Conditioning (Sep- 
tember 1968). 

5.11 Sheet Metal and Air Conditioning Contractors' Na- 


tional Assoc., Inc.: Round Industrial Duct Construc- 
tion Standards. SMACNA, Vienna, VA (1977). 


5.12 Sheet Metal and Air Conditioning Contractors' Na- 


Exhaust System Design Procedure 5-39 


tional Assoc., Inc.: Rectangular Industrial Duct Con- 
struction Standards. SMACNA , Vienna, VA (1980). 


5.13 American Welding Society: AWS DI.1-72. AWS, 
Miami, FL. 


5.14 Air Movement & Control Associations, Inc.: AMCA 
Publication 201. AMCA, Arlington Heights, IL. 


5.15 Wright, Jr., D.K.: A New Friction Chart for Round 
Ducts. ASHVE Transactions, Vol. 51, p. 303 (1945). 


5.16 Clarke, J.H.: Air Flow Around Buildings. Heating, 
Piping and Air Conditioning. 39(5):145-154 (May 
1967). 


5.17 American Society of Heating, Refrigerating and Air- 
Conditioning Engineers: 1989 ASHRAE Handbook, 
Fundamentals Volume, p. 32.31. ASHRAE, Atlanta, 
GA (1989). 


5.18 Brandt, A.D.: Industrial Health Engineering. John 
Wiley & Sons, New York (1947). 


5.19 American Society of Heating, Refrigerating and Air- 
Conditioning Engineers: Heating, Ventilating, Air 
Conditioning Guide, 37th ed. ASHRAE, Atlanta, GA 
(1959). 


5.20 American Society of Heating, Refrigerating and Air 
Conditioning Engineers: 1993 ASHRAE Handbook, 
Fundamentals Volume, Chapter 5. ASHRAE Atlanta, 
GA (1993). 


5-40 Industrial Ventilation 


he= 0.93 VP, he= 6.49 VP, he= 0.04 VP, 
PLAIN DUCT END FLANGED DUCT END BELLMOUTH ENTRY 


45° taper angle =o. c j— 


—_ 


he= 1.5 VP, 
h. = 0.4 VP, (tapered toke-off) TRAP OR SETTLING CHAMBER 


he= 1.78 Woritice “hh. = 0.65 VP, (no taper) 
SHARP—EDGED Spina A ne cies 
ORIFICE STANDARD GRINDER HOOD 


TAPERED HOODS 1.10 
Flanged or unflanged; round, squore or 1.00 [ | | 
rectangular. 8 is the major angle on ENTRY LOSS (hg) 
rectangular hoods. © |ROUND RECTANGULAR 0.90 
0.25 VP = 
0.16 VP 28 
015 VP 2 0.70 Rectengilar & Square | 
0.17 VP i Transition to Round 
0.25 VP G 0.607 
0.55 VP RB 0.507 
0.48 VP Sg : 
0.50 VP 5. 0.40 
= 0.30 s 
VP = Duct VP = VPg ZO at 
Note: 180° values represent mS) 0.20 2 ap Wo poticall 
round ducts butted into 0.10 Kb em Lil sal ie 
back of booth or hood ; or ores Ref. 5-14) 
without a rectangular to : aap are Te : 
Face area (A) at least 2 times the duct area. round transition. 0 20 40 60 80 100 120140 160 180 i 


®, INCLUDED ANGLE IN DEGREES 


MISCELLANEOUS VALUES 


COMPOUND HOODS 


A compound hood, such as the HOOD ENTRY LOSS 

slot/plenum shown to the right, COEFFICIENT Fy 

would have 2 losses, one through Abrasive blast chamber 1.0 

the slot and the other through Abrasive blast elevator 255 

the transition into the duct. Abrasive separator 2.3 
Elevators (enclosures) 0.69 

The slot entry loss coefficient, F,, Flanged pipe plus close elbow 0.8 

would have a value typically in the Plain pipe plus close elbow 1.60 


range of 1.00 to 1.78 (see Chapters 
3 and 10). 


The duct entry loss coefficient is given 
by the above data for tapered hoods. 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


HOOD ENTRY 
LOSS COEEFPICIENTS 


7-95 _ pur “5-7 _ 


Exhaust System Design Procedure 5-41 


Stamped 5S-plece 


(Smooth) 


2.00 [| 


t—piece 
3--piece 


[050 | 0.37 [| 0. 
0.54 | 0.42 4 0.5 


* extrapolated fromm published data 


OTHER ELBOW LOSS COEFFICIENTS 
Mitered, no vones ees 
Mitered, turning vanes 0.6 
Flatback (R/D = 2.5) 0.05 (see Figure 5-23) 


NOTE: Loss factors are assumed to be for elbows of “zero length.” Friction losses should be 
3 
included to ihe intersection of centerlines. 


ROUND ELBOW LOSS COEFFICIENTS 


(Ref. 5.13 


SQUARE & RECTANGULAR ELBOW LOSS COEFFICIENTS 


DUCT DESIGN DATA 
ELBOW LOSSES 


AMERICAN CONFERENCE 
: OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


5-42 Industrial Ventilation 


Angle 0 | Loss Fraction o! 


it Branch 


0. 06 


Note: Branch eniry loss assumed to occur 
in branch and is so calculated. 


Do not include an ee regain 
calculation for branch entry enlargements 


BRANCH ENTRY LOSSES 


mi, No. of 


seo. Fractinn ar We 
iirarieters Loss Fraction of VI 


0.50 D 


0.45 D 


WEATHER CAP 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
eee ee 


DUCT DESIGN DATA 


Table 5-5 Tabulated Friction Loss Factors 
Galvanized Sheet Metal Duct 


Exhaust System Design Procedure 


Diameter Friction Loss, No. VP per foot 
inches 1000 fpm 2000 fpm 3000 fpm 4000 fpm 5000 fpm 6000 fpm 
0.5 1.0086 0.9549 0.9248 0.9040 0.8882 0.8755 
1 0.4318 0.4088 0.3959 0.3870 0.3802 0.3748 
1.5 0.2629 0.2489 0.2410 0.2356 0.2315 0.2282 
2 0.1848 0.1750 0.1695 0.1657 0.1628 0.1605 
2.5 0.1407 0.1332 0.1290 0.1261 0.1239 0.1221 
3 0.1125 0.1065 0.1032 0.1009 0.0991 0.0977 
3.5 0.0932 0.0882 0.0854 0.0835 0.0821 0.0809 
4 0.0791 0.0749 0.0726 0.0709 0.0697 0.0687 
4.5 0.0685 0.0649 0.0628 0.0614 0.0603 0.0595 
5 0.0602 0.0570 0.0552 0.0540 0.0530 0.0523 
§.5 0.0536 0.0507 0.0491 0.0480 0.0472 0.0465 
6 0.0482 0.0456 0.0442 0.0432 0.0424 0.0418 
7 0.0399 0.0378 0.0366 0.0358 0.0351 0.0346 
8 0.0339 0.0321 0.0311 0.0304 0.0298 0.0294 
9 0.0293 0.0278 0.0269 0.0263 0.0258 0.0255 
10 0.0258 0.0244 0.0236 0.0231 0.0227 0.0224 
11 0.0229 0.0217 0.0210 0.0206 0.0202 0.0199 
12 0.0206 0.0195 0.0189 0.0185 0.0182 0.0179 
13 0.0187 0.0177 0.0171 0.0168 0.0165 0.0162 
14 0.0171 0.0162 0.0157 0.0153 0.0150 0.0148 
15 0.0157 0.0149 0.0144 0.0141 0.0138 0.0136 
16 0.0145 0.0137 0.0133 0.0130 0.0128 0.0126 
17 0.0135 0.0127 0.0123 0.0121 0.0119 0.0117 
18 0.0126 0.0119 0.0115 0.0113 0.0111 0.0109 
19 0.0118 0.0111 0.0108 0.0105 0.0103 0.0102 
20 0.0110 0.0104 0.0101 0.0099 0.0097 0.0096 
21 0.0104 0.0098 0.0095 0.0093 0.0092 0.0090 
22 0.0098 0.0093 0.0090 0.0088 0.0086 0.0085 
23 0.0093 0.0088 0.0085 0.0083 0.0082 0.0081 
24 0.0088 0.0084 0.0081 0.0079 0.0078 0.0077 
25 0.0084 0.0080 0.0077 0.0075 0.0074 0.0073 
26 0.0080 0.0076 0.0073 0.0072 0.0070 0.0069 
27 0.0076 0.0072 0.0070 0.0069 0.0067 0.0066 
28 0.0073 0.0069 0.0067 0.0066 0.0064 0.0063 
29 0.0070 0.0066 0.0064 0.0063 0.0062 0.0061 
30 0.0067 0.0064 0.0062 0.0060 0.0059 0.0058 
31 0.0065 0.0061 0.0059 0.0058 0.0057 0.0056 
32 0.0062 0.0059 0.0057 0.0056 0.0055 0.0054 
0.533 
H, = 0.0307 Over 


5-43 


5-44 Industrial Ventilation 


Table 5-5 Tabulated Friction Loss Factors (cont'd) 
Galvanized Sheet Metal Duct 


Diameter Friction Loss, No. VP per foot 
inches 1000 fpm 2000 fpm 3000 fpm 4000 fpm 5000 fpm 6000 fpm 
33 0.0060 0.0057 0.0055 0.0054 0.0053 0.0052 
34 0.0058 0.0055 0.0053 0.0052 0.0051 0.0050 
35 0.0056 0.0053 0.0051 0.0050 0.0049 0.0048 
36 0.0054 0.0051 0.0049 0.0048 0.0047 0.0047 
37 0.0052 0.0049 0.0048 0.0047 0.0046 0.0045 
38 0.0050 0.0048 0.0046 0.0045 0.0044 0.0044 
39 0.0049 0.0046 0.0045 0.0044 0.0043 0.0042 
40 0.0047 0.0045 0.0043 0.0042 0.0042 0.0041 
41 0.0046 0.0043 0.0042 0.0041 0.0040 0.0040 
42 0.0045 0.0042 0.0041 0.0040 0.0039 0.0039 
43 0.0043 0.0041 0.0040 0.0039 0.0038 0.0038 
44 0.0042 0.0040 0.0039 0.0038 0.0037 0.0036 
45 0.0041 0.0039 0.0038 0.0037 0.0036 0.0036 
46 0.0040 0.0038 0.0037 0.0036 0.0035 0.0035 
47 0.0039 0.0037 0.0036 0.0035 0.0034 0.0034 
48 0.0038 0.0036 0.0035 0.0034 0.0033 0.0033 
49 0.0037 0.0035 0.0034 0.0033 0.0032 0.0032 
50 0.0036 0.0034 0.0033 0.0032 0.0032 0.0031 
52 0.0034 0.0032 0.0031 0.0031 0.0030 0.0030 
54 0.0033 0.0031 0.0030 0.0029 0.0029 0.0028 
56 0.0031 0.0030 0.0029 0.0028 0.0028 0.0027 
58 0.0030 0.0028 0.0027 0.0027 0.0026 0.0026 
60 0.0029 0.0027 0.0026 0.0026 0.0025 0.0025 
62 0.0028 0.0026 0.0025 0.0025 0.0024 0.0024 
64 0.0027 0.0025 0.0024 0.0024 0.0023 0.0023 
66 0.0026 0.0024 0.0023 0.0023 0.0023 0.0022 
68 0.0025 0.0023 0.0023 0.0022 0.0022 0.0021 
70 0.0024 0.0023 0.0022 0.0021 0.0021 0.0021 
72 0.0023 0.0022 0.0021 0.0021 0.0020 0.0020 
74 0.0022 0.0021 0.0020 0.0020 0.0020 0.0019 
76 0.0022 0.0020 0.0020 0.0019 0.0019 0.0019 
78 0.0021 0.0020 0.0019 0.0019 0.0018 0.0018 
80 0.0020 0.0019 0.0019 0.0018 0.0018 0.0018 
82 0.0020 0.0019 0.0018 0.0018 0.0017 0.0017 
84 0.0019 0.0018 0.0017 0.0017 0.0017 0.0017 
86 0.0019 0.0018 0.0017 0.0017 0.0016 0.0016 
88 0.0018 0.0017 0.0017 0.0016 0.0016 0.0016 
90 0.0018 0.0017 0.0016 0.0016 0.0015 0.0015 


Exhaust System Design Procedure 5-45 


Table 5-6 Tabulated Friction Loss Factors 
Black Iron, Aluminum, Stainless Steel, PVC Ducts 


Diameter Friction Loss, No. VP per foot 
inches 1000 fpm 2000 fpm 3000 fpm 4000 fpm 5000 fpm 6000 fpm 
0.5 0.8757 0.7963 0.7533 0.7242 0.7024 0.6851 
1 0.3801 0.3457 0.3270 0.3143 0.3049 0.2974 
1.5 0.2333 0.2121 0.2007 0.1929 0.1871 0.1825 
2 0.1650 0.1500 0.1419 0.1364 0.1323 0.1291 
2.5 0.1261 0.1147 0.1085 0.1043 0.1012 0.0987 
3 0.1013 0.0921 0.0871 0.0837 0.0812 0.0792 
3.5 0.0841 0.0765 0.0724 0.0696 0.0675 0.0658 
4 0.0716 0.0651 0.0616 0.0592 0.0574 0.0560 
4.5 0.0621 0.0565 0.0535 0.0514 0.0499 0.0486 
5 0.0547 0.0498 0.0471 0.0453 0.0439 0.0428 
§.5 0.0488 0.0444 0.0420 0.0404 0.0392 0.0382 
6 0.0440 0.0400 0.0378 0.0364 0.0353 0.0344 
7 0.0365 0.0332 0.0314 0.0302 0.0293 0.0286 
8 0.0311 0.0283 0.0267 0.0257 0.0249 0.0243 
9 0.0270 0.0245 0.0232 0.0223 0.0216 0.0211 
10 0.0238 0.0216 0.0204 0.0197 0.0191 0.0186 
11 0.0212 0.0193 0.0182 0.0175 0.0170 0.0166 
12 0.0191 0.0174 0.0164 0.0158 0.0153 0.0149 
13 0.0173 0.0158 0.0149 0.0143 0.0139 0.0136 
14 0.0158 0.0144 0.0136 0.0131 0.0127 0.0124 
15 0.0146 0.0133 0.0125 0.0121 0.0117 0.0114 
16 0.0135 0.0123 0.0116 0.0112 0.0108 0.0106 
17 0.0125 0.0114 0.0108 0.0104 0.0101 0.0098 
18 0.0117 0.0106 0.0101 0.0097 0.0094 0.0092 
19 0.0110 0.0100 0.0094 0.0091 0.0088 0.0086 
20 0.0103 0.0094 0.0089 0.0085 0.0083 0.0081 
21 0.0097 0.0088 0.0084 0.0080 0.0078 0.0076 
22 0.0092 0.0084 0.0079 0.0076 0.0074 0.0072 
23 0.0087 0.0079 0.0075 0.0072 0.0070 0.0068 
24 0.0083 0.0075 0.0071 0.0068 0.0066 0.0065 
25 0.0079 0.0072 0.0068 0.0065 0.0063 0.0062 
26 0.0075 0.0068 0.0065 0.0062 0.0060 0.0059 
27 0.0072 0.0065 0.0062 0.0059 0.0058 0.0056 
28 0.0069 0.0063 0.0059 0.0057 0.0055 0.0054 
29 0.0066 0.0060 0.0057 0.0055 0.0053 0.0052 
30 0.0063 0.0058 0.0054 0.0052 0.0051 0.0050 
31 0.0061 0.0055 0.0052 0.0050 0.0049 0.0048 
32 0.0059 0.0053 0.0050 0.0048 0.0047 0.0046 
0.465 
H, = 0.0425 


ese 


5-46 Industrial Ventilation 


Table 5-6 Tabulated Friction Loss Factors (cont'd) 


Black Iron, Aluminum, Stainless Steel, PVC Ducts 


Diameter Friction Loss, No. VP per foot 
inches 1000 fpm 2000 fpm 3000 fpm 4000 fpm 5000 fpm 6000 fpm 
33 0.0056 0.0051 0.0049 0.0047 0.0045 0.0044 
34 0.0054 0.0050 0.0047 0.0045 0.0044 0.0043 
35 0.0053 0.0048 0.0045 0.0043 0.0042 0.0041 
36 0.0051 0.0046 0.0044 0.0042 0.0041 0.0040 
37 0.0049 0.0045 0.0042 0.0041 0.0039 0.0038 
38 0.0048 0.0043 0.0041 0.0039 0.0038 0.0037 
39 0.0046 0.0042 0.0040 0.0038 0.0037 0.0036 
40 0.0045 0.0041 0.0039 0.0037 0.0036 0.0035 
41 0.0043 0.0040 0.0037 0.0036 0.0035 0.0034 
42 0.0042 0.0038 0.0036 0.0035 0.0034 0.0033 
43 0.0041 0.0037 0.0035 0.0034 0.0033 0.0032 
44 0.0040 0.0036 0.0034 0.0033 0.0032 0.0031 
45 0.0039 0.0035 0.0033 0.0032 0.0031 0.0030 
46 0.0038 0.0034 0.0033 0.0031 0.0030 0.0030 
47 0.0037 0.0034 0.0032 0.0030 0.0030 0.0029 
48 0.0036 0.0033 0.0031 0.0030 0.0029 0.0028 
49 0.0035 0.0032 0.0030 0.0029 0.0028 0.0027 
50 0.0034 0.0031 0.0029 0.0028 0.0027 0.0027 
52 0.0033 0.0030 0.0028 0.0027 0.0026 0.0026 
54 0.0031 0.0028 0.0027 0.0026 0.0025 0.0024 
56 0.0030 0.0027 0.0026 0.0025 0.0024 0.0023 
58 0.0029 0.0026 0.0025 0.0024 0.0023 0.0022 
60 0.0027 0.0025 0.0024 0.0023 0.0022 0.0021 
62 0.0026 0.0024 0.0023 0.0022 0.0021 0.0021 
64 0.0025 0.0023 0.0022 0.0021 0.0020 0.0020 
66 0.0024 0.0022 0.0021 0.0020 0.0020 0.0019 
68 0.0024 0.0021 0.0020 0.0020 0.0019 0.0018 
70 0.0023 0.0021 0.0020 0.0019 0.0018 0.0018 
72 0.0022 0.0020 0.0019 0.0018 0.0018 0.0017 
74 0.0021 0.0019 0.0018 0.0018 0.0017 0.0017 
76 0.0021 0.0019 0.0018 0.0017 0.0017 0.0016 
78 0.0020 0.0018 0.0017 0.0017 0.0016 0.0016 
80 0.0019 0.0018 0.0017 0.0016 0.0016 0.0015 
82 0.0019 0.0017 0.0016 0.0016 0.0015 0.0015 
84 0.0018 0.0017 0.0016 0.0015 0.0015 0.0014 
86 0.0018 0.0016 0.0015 0.0015 0.0014 0.0014 
88 0.0017 0.0016 0.0015 0.0014 0.0014 0.0014 
90 0.0017 0.0015 0.0015 0.0014 0.0014 0.0013 
0.465 
H, = 0.0425 


GP see 


Exhaust System Design Procedure 5-47 


PSSURE REGAINS FOR EXE PAN SIONS - 


4 2 min 


With ‘in. duct 


“of ‘inlet 


terms lengl int 
tO 4 diam 


t 


Sioa O/o| 


ie SP: = = SP, = 


Abrupt 90] 2 0 ] 
Where: SP, = SR + ROP, - We, oo ‘When SP, =0 (atmosphere) a 


The regain (R 2) will only be 70% ‘of volue shown above when expansion follows a disturbance 
elbow (including a fan) by less than 5 duct diameters. 


STATIC PRESSURE LOSSES FOR CONTRACTIONS 


Abrupi ale 
SP, = SP, -(VP, ~ 


Se I icra 


Ratio M/A, | 


i 


Ol 


jon 


Oo 


Abrupt contraction 


duct area, 
Note: 
in caiculating SP for expansion or contraction use algebraic signs: VP is (+), and usually 
SP is (+) in discharge duct from fon, and SP is (-) in inlet duct to fon. 
AME RICAN CONFERENCE 
OF GOVERNMENTAL 
es eae HYGIENISTS 


‘Yricure = =5-7G 


5-48 Industrial Ventilation 


TABLE 5-7A. Velocity Pressure to Velocity Conversion — Standard Air 


FROM: V = 4005 V VP V = Velocity, fpm 
VP = Velocity Pressure, “wa 
vP Vv vP v vP Vv vP v vP Vv vP Vv 

0.01 401 0.51 2860 1.01 4025 1.51 4921 2.01 5678 2.60 6458 
0.02 566 0.52 2888 1.02 4045 1.52 4938 2.02 5692 2.70 6581 
0.03 694 0.53 2916 1.03 4065 1.53 4954 2.03 5706 2.80 6702 
0.04 801 0.54 2943 1.04 4084 1.54 4970 2.04 5720 2.90 6820 
0.05 896 0.55 2970 1.05 4104 1.55 4986 2.05 5734 3.00 6937 
0.06 981 0.56 2997 1.06 4123 1.56 5002 2.06 5748 3.10 7052 
0.07 1060 0.57 3024 1.07 4143 1.57 5018 2.07 5762 3.20 7164 
0.08 1133 0.58 3050 1.08 4162 1.58 5034 2.08 5776 3.30 7275 
0.09 1201 0.59 3076 1.09 4181 1.59 5050 2.09 5790 3.40 7385 
0.10 1266 0.60 3102 1.10 4200 1.60 5066 2.10 5804 3.50 7493 
0.11 1328 0.61 3128 1.11 4220 1.61 5082 2.11 5818 3.60 7599 
0.12 1387 0.62 3154 1.12 4238 1.62 5098 2.12 5831 3.70 7704 
0.13 1444 0.63 3179 1.13 4257 1.63 5113 2.13 5845 3.80 7807 
0.14 1499 0.64 3204 1.14 4276 1.64 5129 2.14 5859 3.90 7909 
0.15 1551 0.65 3229 1.15 4295 1.65 5145 2.15 5872 4.00 8010 
0.16 1602 0.66 3254 1.16 4314 1.66 5160 2.16 5886 4.10 8110 
0.17 1651 0.67 3278 1.17 4332 1.67 5176 2.17 5900 4.20 8208 
0.18 1699 0.68 3303 1.18 4351 1.68 5191 2.18 5913 4.30 8305 
0.19 1746 0.69 3327 1.19 4369 1.69 5206 2.19 5927 4.40 8401 
0.20 1791 0.70 3351 1.20 4387 1.70 5222 2.20 5940 4.50 8496 
0.21 1835 0.71 3375 1.21 4405 1.71 5237 2.21 5954 4.60 8590 
0.22 1879 0.72 3398 1.22 4424 1.72 5253 2.22 5967 4.70 8683 
0.23 1921 0.73 3422 1.23 4442 1.73 5268 2.23 5981 4.80 8775 
0.24 1962 0.74 3445 1.24 4460 1.74 5283 2.24 5994 4.90 8865 
0.25 2003 0.75 3468 1.25 4478 1.75 5298 2.25 6007 5.00 8955 
0.26 2042 0.76 3491 1.26 4496 1.76 5313 2.26 6021 5.50 9393 
0.27 2081 0.77 3514 1.27 4513 1.77 5328 2.27 6034 6.00 9810 
0.28 2119 0.78 3537 1.28 4531 1.78 5343 2.28 6047 6.50 10211 
0.29 2157 0.79 3560 1.29 4549 1.79 5358 2.29 6061 7.00 10596 
0.30 2194 0.80 3582 1.30 4566 1.80 5373 2.30 6074 7.50 10968 
0.31 2230 0.81 3604 1.31 4584 1.81 5388 2.31 6087 8.00 11328 
0.32 2266 0.82 3627 1.32 4601 1.82 5403 2.32 6100 8.50 11676 
0.33 2301 0.83 3649 1.33 4619 1.83 5418 2.33 6113 9.00 12015 
0.34 2335 0.84 3671 1.34 4636 1.84 5433 2.34 6126 9.50 12344 
0.35 2369 0.85 3692 1.35 4653 1.85 5447 2.35 6140 10.00 12665 
0.36 2403 0.86 3714 1.36 4671 1.86 5462 2.36 6153 10.50 12978 
0.37 2436 0.87 3736 1.37 4688 1.87 5477 2.37 6166 11.00 13283 
0.38 2469 0.88 3757 1.38 4705 1.88 5491 2.38 6179 11.50 13582 
0.39 2501 0.89 3778 1.39 4722 1.89 5506 2.39 6192 12.00 13874 
0.40 2533 0.90 3799 1.40 4739 1.90 5521 2.40 6205 12.50 14160 
0.41 2564 0.91 3821 1.44 4756 1.91 5535 2.41 6217 13.00 14440 
0.42 2596 0.92 3841 1.42 4773 1.92 5549 2.42 6230 13.50 14715 
0.43 2626 0.93 3862 1.43 4789 1.93 5564 2.43 6243 14.00 14985 
0.44 2657 0.94 3883 1.44 4806 1.94 5578 2.44 6256 14.50 15251 
0.45 2687 0.95 3904 1.45 4823 1.95 5593 2.45 6269 15.00 15511 
0.46 2716 0.96 3924 1.46 4839 1.96 5607 2.46 6282 15.50 15768 
0.47 2746 0.97 3944 1.47 4856 1.97 5621 2.47 6294 16.00 16020 
0.48 2775 0.98 3965 1.48 4872 1.98 5636 2.48 6307 16.50 16268 
0.49 2803 0.99 3985 1.49 4889 1.99 5650 2.49 6320 17.00 16513 


0.50 2832 1.00 4005 1.50 4905 2.00 5664 2.50 6332 17.50 16754 


Exhaust System Design Procedure 5-49 


TABLE 5-78. Velocity to Velocity Pressure Conversion — Standard Air 


FROM: V = 4005 V VP V = Velocity, fpm 
VP = Velocity Pressure, "wg 


5-50 


Industrial Ventilation 


TABLE 5-8. Area and Circumference of Circles 


Inches 


1 
1.5 
2 
2.5 
3 
3.5 
4 
45 
5 
5.5 
6 
6.5 


AREA 
Square Square 
Inches Feet 

0.79 0.0055 
1.77 0.0123 
3.14 0.0218 
4.91 0.0341 
7.07 0.0491 
9.62 0.0668 
12.57 0.0873 
15.90 0.1104 
19.63 0.1364 
23.76 0.1650 
28.27 0.1963 
33.18 0.2304 
38.48 0.2673 
44.18 0.3068 
50.27 0.3491 
56.75 0.3941 
63.62 0.4418 
70.88 0.4922 
78.54 0.5454 
86.59 0.6013 
95.03 0.6600 
103.87 0.7213 
113.10 0.7854 
132.73 0.9218 
153.94 1.0690 
176.71 1.2272 
201.06 1.3963 
226.98 1.5763 
254.47 1.7671 
283.53 1.9689 
314.16 2.1817 
346.36 2.4053 
380.13 2.6398 
415.48 2.8852 
452.39 3.1416 
490.87 3.4088 
530.93 3.6870 
572.56 3.9761 
615.75 4.2761 
660.52 4.5869 


CIRCUMFERENCE 
inches Feet 
3.14 0.2618 
4.71 0.3927 
6.28 0.5236 
7.85 0.6545 
9.42 0.7854 
11.00 0.9163 
12.57 1.0472 
14.14 1.1781 
15.71 1.3090 
17.28 1.4399 
18.85 1.5708 
20.42 1.7017 
21.99 1.8326 
23.56 1.9635 
25.13 2.0944 
26.70 2.2253 
28.27 2.3562 
29.85 2.4871 
31.42 2.6180 
32.99 2.7489 
34.56 2.8798 
36.13 3.0107 
37.70 3.1416 
40.84 3.4034 
43.98 3.6652 
47.12 3.9270 
50.27 4.1888 
53.41 4.4506 
56.55 4.7124 
59.69 4.9742 
62.83 5.2360 
65.97 5.4978 
69.12 5.7596 
72.26 6.0214 
75.40 6.2832 
78.54 6.5450 
81.68 6.8068 
84.82 7.0686 
87.96 7.3304 
91.11 7.5922 


Diam. AREA CIRCUMFERENCE 
in Square Square 
Inches inches Feet Inches Feet 

30 706.9 4.909 94.2 7.854 
31 754.8 5.241 97.4 8.116 
32 804.2 5.585 100.5 8.378 
33 855.3 5.940 103.7 8.639 
34 907.9 6.305 106.8 8.901 
35 962.1 6.681 110.0 9.163 
36 1017.9 7.069 113.1 9.425 
37 1075.2 7.467 116.2 9.687 
38 1134.1 7.876 119.4 9.948 
39 1194.6 8.296 122.5 10.210 
40 1256.6 8.727 125.7 10.472 
41 1320.3 9.168 128.8 10.734 
42 1385.4 9.621 131.9 10.996 
43 1452.2 10.085 135.1 11.257 
44 1520.5 10.559 138.2 11.519 
45 1590.4 11.045 141.4 11.781 
46 1661.9 11.544 144.5 12.043 
47 1734.9 12.048 147.7 12.305 
48 1809.6 12.566 150.8 12.566 
49 1885.7 13.095 153.9 12.828 
50 1963.5 13.635 157.1 13.090 
52 2123.7 14.748 163.4 13.614 
54 2290.2 15.904 169.6 14.137 
56 2463.0 17.104 175.9 14.661 
58 2642.1 18.348 182.2 15.184 
60 2827.4 19.635 188.5 15.708 
62 3019.1 20.966 194.8 16.232 
64 3217.0 22.340 201.1 16.755 
66 3421.2 23.758 207.3 17.279 
68 3631.7 25.220 213.6 17.802 
70 3848.5 26.725 219.9 18.326 
72 4071.5 28.274 226.2 18.850 
74 4300.8 29.867 232.5 19.373 
76 4536.5 31.503 238.8 19.897 
78 4778.4 33.183 245.0 20.420 
80 5026.5 34.907 251.3 20.944 
82 5281.0 36.674 257.6 21.468 
84 5541.8 38.485 263.9 21.991 
86 5808.8 40.339 270.2 22.515 
88 6082.1 42.237 276.5 23.038 


The usual sheet metal fabricator will have patterns for ducts in 0.5-inch steps through 5.5-inch diameter; 1 inch steps 6 inches through 20 inches 
and 2-inch steps 22 inches and larger diameters. 


Exhaust System Design Procedure 


TABLE 5-9. Circular Equivalents of Rectangular Duct Sizes 


a\°/4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10.5 11.0 11.5 12.0 12.5 13.0 13.5 14.0 14.5 15.0 


15.5 16.0 


3.0 |3.8 4.0 4.2 44 46 4.749 5152535556 57 59 60 61 62 63 64 65 66 67 68 
3.5 14.1 4.3 4.6 48 5.0 5.25355 57586061 63 64 65 67 68 69 70 71 72 73 75 
4.0 14.446 495153555759 61636466 67 69 70 72 73 74 76 77 78 79 80 
4.5 (44649 52545759616365676970 72 74 75 77 78 7.9 81 82 84 85 86 
5.0 14.9 5.2 5.5 5.7 6.0 6.2 6.4 6.7 6.9 7.1 7.3 7.4 78 80 81 83 84 86 87 89 90 91 
5.5 6.1 5.4 5.7 6.0 6.3 6.5 6.8 7.0 7.2 7.4 7.6 7.8 8.2 84 86 87 89 90 92 93 95 9.6 


6.9 
7.6 
8.2 
8.7 
9.3 
9.8 


7.0 
77 
8.3 
8.8 
9.4 
9.9 


a\*] 6.0 7.0 8.0 9.0 10.0 11.0 12.0 13.0 14.0 15.0 16.0 17.0 18.0 19.0 20.0 22.0 24.0 26.0 28.0 30.0 32.0 34.0 36.0 38.0 40.0 


6.0) 6.6 

7.0) 7.1 7.7 

8.0) 7.6 82 8.7 

9.0) 80 87 93 9.8 

10.0} 84 9.1 9.8 10.4 10.9 

11.0) 8.8 9.5 10.2 10.9 11.5 12.0 

12.0) 9.1 9.9 10.7 11.3 12.0 12.6 13.1 

13.0) 9.5 10.3 11.1 11.8 12.4 13.1 13.7 14.2 

14.0) 9.8 10.7 11.5 12.2 12.9 13.5 14.2 14.7 15.3 

15.0/10.1 11.0 11.8 12.6 13.3 14.0 14.6 15.3 15.8 16.4 

16.0}10.4 11.3 12.2 13.0 13.7 14.4 15.1 15.7 16.4 16.9 17.5 

17.0/10.7 11.6 12.5 13.4 14.1 14.9 15.6 16.2 16.8 17.4 18.0 18.6 

18.0/11.0 11.9 12.9 13.7 14.5 15.3 16.0 16.7 17.3 17.9 18.5 19.1 19.7 

19.0)11.2 12.2 13.2 14.1 14.9 15.7 16.4 17.1 17.8 18.4 19.0 19.6 20.2 20.8 

20.0/11.5 12.5 13.5 14.4 15.2 16.0 16.8 17.5 18.2 18.9 19.5 20.1 20.7 21.3 21.9 

22.0)12.0 13.0 14.1 15.0 15.9 16.8 17.6 18.3 19.1 19.8 20.4 21.1 21.7 22.3 22.9 24.0 

24.0/12.4 13.5 14.6 15.6 16.5 17.4 18.3 19.1 19.9 20.6 21.3 22.0 22.7 23.3 23.9 25.1 26.2 

26.0/12.8 14.0 15.1 16.2 17.1 18.1 19.0 19.8 20.6 21.4 22.1 22.9 23.5 24.2 24.9 26.1 27.3 28.4 

28.0/13.2 14.5 15.6 16.7 17.7 18.7 19.6 20.5 21.3 22.1 22.9 23.7 24.4 25.1 25.8 27.1 28.3 29.5 30.6 

30.0)13.6 14.9 16.1 17.2 18.3 19.3 20.2 21.1 22.0 22.9 23.7 24.4 25.2 25.9 26.6 28.0 29.3 30.5 31.7 32.8 

32.0/14.0 15.3 16.5 17.7 18.8 19.8 20.8 21.8 22.7 23.5 24.4 25.2 26.0 26.7 27.5 28.9 30.2 31.5 32.7 33.9 35.0 
34.0)14.4 15.7 17.0 18.2 19.3 20.4 21.4 22.4 23.3 24.2 25.1 25.9 26.7 27.5 28.3 29.7 31.1 32.4 33.7 34.9 36.1 37.2 
36.0/14.7 16.1 17.4 18.6 19.8 20.9 21.9 22.9 23.9 24.8 25.7 26.6 27.4 28.2 29.0 30.5 32.0 33.3 34.6 35.9 37.1 38.2 39.4 
38.0)15.0 16.5 17.8 19.0 20.2 21.4 22.4 23.5 24.5 25.4 26.4 27.2 28.1 28.9 29.8 31.3 32.8 34.2 35.6 36.8 38.1 39.3 40.4 41.5 
40.0)15.3 16.8 18.2 19.5 20.7 21.8 22.9 24.0 25.0 26.0 27.0 27.9 28.8 29.6 30.5 32.1 33.6 35.1 36.4 37.8 39.0 40.3 41.5 42.6 43.7 
42.0/15.6 17.1 18.5 19.9 21.1 22.3 23.4 24.5 25.6 26.6 27.6 28.5 29.4 30.3 31.2 32.8 34.4 35.9 37.3 38.7 40.0 41.3 42.5 43.7 44.8 
44.0115.9 17.5 18.9 20.3 21.5 22.7 23.9 25.0 26.1 27.1 28.1 29.1 30.0 30.9 31.8 33.5 35.1 36.7 38.1 39.5 40.9 42.2 43.5 44.7 45.8 
46.0}16.2 17.8 19.3 20.6 21.9 23.2 24.4 25.5 26.6 27.7 28.7 29.7 30.6 31.6 32.5 34.2 35.9 37.4 38.9 40.4 41.8 43.1 44.4 45.7 46.9 
48.0}16.5 18.1 19.6 21.0 22.3 23.6 24.8 26.0 27.1 28.2 29.2 30.2 31.2 32.2 33.1 34.9 36.6 38.2 39.7 41.2 42.6 44.0 45.3 46.6 47.9 
50.0)16.8 18.4 19.9 21.4 22.7 24.0 25.2 26.4 27.6 28.7 29.8 30.8 31.8 32.8 33.7 35.5 37.2 38.9 40.5 42.0 43.5 44.9 46.2 47.5 48.8 
54.0)17.3 19.0 20.6 22.0 23.5 24.8 26.1 27.3 28.5 29.7 30.8 31.8 32.9 33.9 34.9 36.8 38.6 40.3 41.9 43.5 45.1 46.5 48.0 49.3 50.7 
58.0)17.8 19.5 21.2 22.7 24.2 25.5 26.9 28.2 29.4 30.6 31.7 32.8 33.9 35.0 36.0 38.0 39.8 41.6 43.3 45.0 46.6 48.1 49.6 51.0 52.4 
62.0/18.3 20.1 21.7 23.3 24.8 26.3 27.6 28.9 30.2 31.5 32.6 33.8 34.9 36.0 37.1 39.1 41.0 42.9 44.7 46.4 48.0 49.6 51.2 52.7 54.1 
66.0)18.8 20.6 22.3 23.9 25.5 26.9 28.4 29.7 31.0 32.3 33.5 34.7 35.9 37.0 38.1 40.2 42.2 44.1 46.0 47.7 49.4 51.1 52.7 54.2 55.7 
70.0)/19.2 21.1 22.8 24.5 26.1 27.6 29.1 30.4 31.8 33.1 34.4 35.6 36.8 37.9 39.1 41.2 43.3 45.3 47.2 49.0 50.8 52.5 54.1 55.7 57.3 
74.0/19.6 21.5 23.3 25.1 26.7 28.2 29.7 31.2 32.5 33.9 35.2 36.4 37.7 38.8 40.0 42.2 44.4 46.4 48.4 50.3 52.1 53.8 55.5 57.2 58.8 
78.0)20.0 22.0 23.8 25.6 27.3 28.8 30.4 31.8 33.3 34.6 36.0 37.2 38.5 39.7 40.9 43.2 45.4 47.5 49.5 51.4 53.3 55.1 56.9 58.6 60.2 
82.0/20.4 22.4 24.3 26.1 27.8 29.4 31.0 32.5 33.9 35.4 36.7 38.0 39.3 40.6 41.8 44.1 46.4 48.5 50.6 52.6 54.5 56.4 58.2 59.9 61.6 
86.0/20.8 22.9 24.8 26.6 28.3 30.0 31.6 33.1 34.6 36.1 37.4 38.8 40.1 41.4 42.6 45.0 47.3 49.6 51.7 53.7 55.7 57.6 59.4 61.2 63.0 
90.0/21.2 23.3 25.2 27.1 28.9 30.6 32.2 33.8 35.3 36.7 38.2 39.5 40.9 42.2 43.5 45.9 48.3 50.5 52.7 54.8 56.8 58.8 60.7 62.5 64.3 


5-52 Industrial Ventilation 


TABLE 5-9. Circular Equivalents of Rectangular Duct Sizes (cont.) 


42.0 44.0 46.0 48.0 50.0 54.0 58.0 62.0 66.0 70.0 74.0 78.0 82.0 86.0 90.0 


(A x B)°-625 
“(A+ BPs 


equiv = equivalent round duct size for rectangular 
duct, in. 


24.0 
26.0 A = one side of rectangular duct, in. 
28.0 B = adjacent side of rectangular duct, in. 


44.0/47.0 48.1 

46.0 }48.0 49.2 50.3 

48.0 /49.1 50.2 51.4 52.5 

50.0/50.0 51.2 52.4 53.6 54.7 

54.0)52.0 53.2 54.4 55.6 56.8 59.0 

98.0 53.8 55.1 56.4 57.6 58.8 61.2 63.4 

62.0)55.5 56.9 58.2 59.5 60.8 63.2 65.5 67.8 

66.0|57.2 58.6 60.0 61.3 62.6 65.2 67.6 69.9 72.1 

70.0 }58.8 60.3 61.7 63.1 64.4 67.1 69.6 72.0 74.3 76.5 

74.0}60.3 61.9 63.3 64.8 66.2 68.9 71.5 74.0 76.4 78.7 80.9 

78.0)/61.8 63.4 64.9 66.4 67.9 70.6 73.3 75.9 78.4 80.7 83.0 85.3 

82.0/63.3 64.9 66.5 68.0 69.5 72.3 75.1 77.8 80.3 82.8 85.1 87.4 89.6 
86.0 64.7 66.3 67.9 69.5 71.0 74.0 76.8 79.6 82.2 84.7 87.1 89.5 91.8 94.0 
90.0 |66.0 67.7 69.4 71.0 72.6 75.6 78.5 81.3 84.0 86.6 89.1 91.5 93.9 95.2 98.4 


TABLE 5-10. Air Density Correction Factor, df 


Exhaust System Design Procedure 5-53 


— 5000 - 4000 —3000 —2000 — 1000 


ALTITUDE RELATIVE TO SEA LEVEL, ft 
0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 


"Hg 35.74 34.51 
“w 486.74 469.97 


33.31 
453.67 


32.15 
437.84 


31.02 
422.45 


Barometric Pressure 
29.92 28.86 27.82 26.82 25.84 24.89 23.98 23.09 22.22 21.39 20.57 
407.50 392.98 378.89 365.21 351.93 339.04 326.54 314.42 302.66 291.26 280.21 


Density Factor, df 


1.22 
1.11 
1.02 
0.96 
0.91 
0.84 
0.77 
0.72 
0.67 
0.63 
0.59 
0.56 
0.53 
0.51 
0.48 
0.44 
0.41 
0.38 
0.35 


1.17 
1.07 
0.99 
0.93 
0.88 
0.81 
0.75 
0.69 
0.65 
0.61 
0.57 
0.54 
0.51 
0.49 
0.46 
0.42 
0.39 
0.36 
0.34 


1.13 
1.03 
0.95 
0.90 
0.85 
0.78 
0.72 
0.67 
0.62 
0.59 
0.55 
0.52 
0.49 
0.47 
0.45 
0.41 

0.38 
0.35 
0.33 


Industrial Ventilation 


5-54 


es ee re <2 aH aias8§ 


ayy Aig 40 punog jad sulcus - juayzuo> ainysiow 


La 


Z 
pa Le 
= 


5 
5 = 
ai Ls F F an Saar Chara 
bt : 
A 


NN 


-siese 


ez 


NTN 
UTNE 


= 
P| 
a: 
—_ 
CZ) 
ae 

wr 
“A 248 


ee A, 
x S. 


MT 
V 
oe, 


a NRDSRNANN, 


NY 

ane 
NNT 

Na ePetT NALINI 


aie 
NV 


AST Re wie A 
vab'a 
AN ANZ We a PO 
LAMA AVA I 


PSYCHROMETRIC CHART 
Barometric Pressure 29.92" Hg 
Na 
NJ 
IS 


Total Heat Values - ASHAE Guide 
End Points - Zimmerman & Lavine 


FIGURE 5-17 


1959 


© American Ate Filter Co. tnc., 
Louisville, Ky. Form 1932 


5-55 


Exhaust System Design Procedure 


of | J 4 
© — 
that 


HIGH TEMPERATURES 


Garomotric Prossure 


al 
os 
g 
= 
w 
= 
‘es 
-— 
tna 
& 
i=] 
ot 
= 
wi 
> 
wa 
Bo 


29.92 In. Hg 


2 A a A 
CTO TSS OEE OT & 
: 


fo) 


OP. Gs O88 OE Va 9 ee ees id re 
WAG ww steeds, ke ee, eee 


if 
i 


% 
(848 es A ee 


FIGURE 5-18 


140 «150 160 17 
DRY BULB TEMPERATURE F 


eee ee 28 EE A By on 
eT Oe AE EE ae oS SY AP a nt Oe en 5 


Te A BE te ORY: EA ED 
a AS A POE SED OF ABS Pt a A es: SUR a 


Barometric Pressure 29,92 in ae: 
vs 


a 


Vale esac 
isn lie aaey 
eT 


Rie. 
Thea Soares 


KOEN SARS SSS 
Sit SSeS 


= 
P—tN_ 
Le 
Bs Ga 
ea 
avi 
a 
L_ | 


i is 
| 
cars 


ioe 
LA 
AZ 
i 
ie) 
Vy, 
z 


PLL YA 
ria WAVE Za 
AAS 


Ks 


U 


ee 


j 
a 


7 rT VT 
ae Cy, 


me 
Zw 7s 
vmiaien 


Oe SS 
PAL LAT \ 


DRY BULB TEMPERATURE IN DEGREES F. 


p 


Lt 
|_| 


B 


8 
a 
8 i 
3 


FIGURE 5-19. Psychrometric chart for humid air based on one pound dry weight (© 1951 American Air Filter Co., Inc., Louisville, KY) 


9S-S 


uoOHeQUIA [BLySsnpUy 


POUNDS OF WATER PER POUND OF DRY AIR 


BASED ON ONE POUND DRY WEIGHT 


oe 
< 
a 
= 
=) 
EE 
5 
Te 
oe 
a 
< 
rc 
O 
YQ 
ac 
Ee 
uJ 
S 
a 
5 
> 
Vv) 
a 


ye: 


Ca 


COPYRIGHT 193! 
AMERICAN AIR FILTER CO.INC. 


HUMID VOLUME — CU, FT./LB. ORY AIR 


DENSITY FACTOR - MIXTURE 


LOUISVILLE, KY. 
Barometric Pressure 29.92 in Hg 


emcees 
APTI eet 
fig piaralee er at 
HET Lee 


Exhaust System Design Procedure 


- &, 


ae 3 
jong 


1B... 


DEGREES F. 


TEMPERATURE 


ORY BULB 


5-57 


FIGURE 5-20 


5-58 Industrial Ventilation 


i 
ike fom 


S—- 2 to 2.5 dia. ee 5 dia. 
center line ng CLR. 
radius (C.LR.) 


ACCEPTABLE AVOID 
ELBOW RADIUS 
[lbows should be 2 to 2.5 diameter centerline radius excest 


wnere space does not permit. See Fig. 5-13 for loss factor. 


|<< W — Pa 


H 
i 


PREFERRED 
ASPECT RATIO (fH) 


\ ~ 
Elbows should have (8) and @ equal to or greater than 
See Fig. 5-13 for loss factor. 


Note: Avoid mitered elbows. If necessary, use only with clean 
air and provide turning vanes. Consult mfg. for turning 
vane loss factor. 


| AMERICAN CONFERENCE | PRINCIPLES OF DUCT DESIGN | 
! OF GOVERNMENTAL E L BOM Ws : 
a INDUSTRIAL HYGIENISTS B= 


Exhaust System Design Procedure 5-59 


~ RUBBER 
BEL TING 


AIR FLOW 


Ne REMOVABLE 
WEAR PLATE 


ANGLE IRON 


REMOVABLE WEAR PLATE 
12 ga. OR HEAVIER. 


FLAT BACK 


a> 3" MINIMUM 
CONCRETE 


cE [e [3 C) W 


NOTE: PROVIDE SOLID MOUNTING FOR CONCRETE REINFORCED ELBOWS. 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
PINDUSTRIAL HYGIENISTS By 


HEAVY DUTY ELBOWS 


5-60 Industrial Ventilation 


aa 


Sian 


Pe iit SG 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


Exhaust System Design Procedure 5-61 


“& DRILL AND RIVET OR BOLT - 
AT FIXED POSITION. 


| AMERICAN CONFERENCE 
} OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


BLAST GATES 


5-62 Industrial Ventilation 


DUCT ENLARGEMENTS 


! See Fig. 5-16 


PREFERRED AVOID 


DUCT CONTRACTIONS 


/ See Fig. 5-16 


PREFERRED AVOID 


SYMMETRICAL WYES 


30 
' 30 AV 

60 ig - 
PREFERRED PREFERRED 


| AMERICAN CONFERENCE | ny | 
ie Coven tad (PRINCIPLES OF DUCT DESIGN 


| INDUSTRIAL HYGIENISTS [y—7—95—— rere —5 25 


Exhaust System Design Procedure 


a 


oe FERRED 


on 
PREFERRED ACCEPTABLE AVOID 
BRANCH ENTRY 
Bronches snould enter ot gradual expansions and oat an angle 


Ley 


of 50° or less (preferred) to 45°if necessary. Expansion should 
be 15°maximum. See Fig. 5-15 for loss coefficients. 


a= Ay t+ Ay k 2O% 


i 
a 
No \_ 
hens \/ . . 1 . 

VM Vern = Minimum transport velocity 


PREFERRED ° > 7 Semen AVOID 
PROPER DUCT SIZE 


Size the duct to maintain the selected or higher 
transport velocity. 


5-63 


| AMERICAN CONFERENCE | PRINCIPLES OF DUCT DESIGN | 
| OF GOVERNMENTAL BRANCH ENTRY 
| INDUSTRIAL HYGIENISTS 


e795 mee 


5-64 Industrial Ventilation 


r~ Tapered inlet 


i aa : 3 ; 
¢ Straight intet 


PREFERRED 


A=twice wheel dia min. 
B=twice wheel dia min. 
C=wheel wicth min. 


ACCEPTABLE AC C| EPTABLE 


Inlet elbow see note inlet eloow see nale 


Note 
See Chapter 6 for system effect 
factors based on inlet and 
outlet duct arrangements. 


Use duct turn vanes to eliminate air 
spin or uneven loading of fan wheel. 


| AMERICAN CONFERENCE | PRINCIPLES OF DUCT DESIGN | 
| OF GOVERNMENTAL — | FAN INLETS 
| INDUSTRIAL HYGIENISTS fm : per rnannrr 


Exhaust System Design Procedure 5-65 


Streamlines 


H | Wind at 
roof heigh 


Vent 


= Forigaaion 


zone 


(oe LULL LL EL 


FIZILIL, 


WZ UMMUITLLLLEIDTTELP 


Zone of 
recirculating flow 


A: Centerline flow patterns around a 
rectangular building 


Undisturbed flow 
we Z1 Roof recirculation region 
~~ 


Z2 High turbulence region 
Z3 Roof Wake boundry 


Building wake 
10:1 oo recirculation 
region 


H 


LZLLL 


So 


LLTIZIDIZ IZZIE ELM ODIO — ZZZZZZEN 
L El 


aaiad Lg oa 


B: Building Recirculation Cavities 


AMERICAN CONFERENCE AIRFLOW AROUND 
| OF GOVERNMENTAL BUILDINGS 


| INDUSTRIAL HYGIENISTS [gg —s—gq aE -| 


5-66 Industrial Ventilation 


7 
Raise due to 
momentum and 
bouyancy 
Effective stack 
heig 


A: Effective stack height 


From weather data: 
Days with max. wind vel. <9 m/s= 98% 
Avg wind vel.= 4.5 m/s 

oN 


For Design: 
Assume 9 m/s, R=1.5, 
then stack vel. should 


Downwash 


Extensive Downwash int 
wake of stack 


° 


Trailing Vortices 


B: Woke Downwash 


| AMERICAN CONFERENCE | EFFECTIVE STACK HEIGHT | 
| OF GOVERNMENTAL 


AND WAKE DOWNWASH | 
) INDUSTRIAL HYGIENISTS [ang _oq 4-94. Jrcuns 5g “ 


Exhaust System Design Procedure 5-67 


Section A-A 
1/20 


iy 4 


Drain) < 


{ 
| 
| 
Bracket upper 
| 
| 
| 


stack to 
bo} += discharge duct come 
| (87/116) (106) 
VERTICAL DISCHARGE OFFSET ELBOWS OFFSET sTack | 


NO LOSS CALCULATE LOSSES DUE TO ELBOWS 


j. Rain protection characteristics of these caps are superior to a deflecting cap located 
0.75D from top of a stack. 


f 2. The length of upper stack is related to rain protection. Excessive additional distance may 


"plowout” of effluent at the gap between upper and lower sections. (86) 
STACKHEAD WEATHER CAP 
oi WZ Equal velocity contours 
= 4 
7 60 10 
3 2 
2 ' : = 
< 75 8 2 2 0 ay 
5 8 ce 50 
7 5 25 loo 
ae 4 a oO 
i. 6 
B15 10 8 6 4 2 = 
100 ; 0 Diameters 
aN NOT RECOMMENDED 
PREFERRED Deflects air downward 


AVOID / 


Delieors air pwn 


| AMERICAN CONFERENCE 
| OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS pate 7-95 


STACKHEAD DESIGNS 


Chapter 6 


FANS 


6.1 INTRODUCTION .................... 6-2 6:3:5> “Ran Laws:s +s 24 seek Meer s ee al aed Shs 6-17 
6.2. BASIC DEFINITIONS ................. 6-2 6.3.6 The Effect of Changing Rotation Rate or 
6.2.1 Ejectors .......020020..000008 | 6-2 Gas DEUSMY cee 4 Geae Se eS 6-17 
62.2 AxialFans.................2.22. 6-2 6.3.7 Limitations on the Use of Fan Laws... . . . 6-19 
6.2.3 CentrifugalFans ................ 6-2 6.3.8 Fan Selection at Air Density Other Than 
6.2.4 Special TypeFans.............0-- 62 Standard 6... ee eee 6-19 
63  FANSELECTION.................... 6-6 6.3.9 Explosive or Flammable Materials ...... 6-21 
6.3.1. Consideration for Fan Selection ........ 6-6 64 FAN aetna AND MAINTENANCE . . . 6-21 
6.3.2 Rating Tables... .........0... 6-15 GA) SystemBleeh, sae yon ehh ys ore 
6.3.3 PointofOperation .............. 6-16 6.4.2 Inspection and Maintenance... ....... 6-25 
6.3.4 Matching Fan Performance REFERENCES... 22 ee 6-25 
and System Requirements .......... 6-17 
Figure 6-1 AirEjectors 2... 0.0. ee eee eee 6-3 Figure 6-15 Fan Discharge Conditions... ......... 6-24 
Figure 6-2 Terminology for Axial and Tubular Centrifugal Figure 6-16 InletElbow ................... 6-25 
Fansrocirs 6 a Sec os Seo ats, athe at te od Ste 6-4 Figure 6-17 Machinery Vibration Severity Chart... .. . 6-26 
Figure 6-3 Terminology for Centrifugal Fan Components . . 6-5 Figure 6-18 | System Effect Curves for Outlet Ducts— 
Figure 6-4 Types of Fans: Impeller and Housing Designs/ Centrifugal Fans .............0.0. 6-27 
Performance Characteristics and Applications . . 6-6 Figure 6-19 System Effect Curves for Outlet Ducts— 
Figure 6-5a Drive Arrangements for Centrifugal Fans 6-11 Axial Fans: og 25 4 ka ee eet eben 6-28 
Figure 6-5b Drive Arrangements for Centrifugal Fans 6-12 Figure 6-20 System Effect Curves for Outlet Elbows 
Figure 6-Sc — Drive Arrangements for Axial Fans with on Centrifugal Fans... 2... 2 ee, 6-29 
or Without Evasé and Inlet Box ... 2... .. 6-13 Figure 6-21 System Effect Curves for Various Mitered Elbows 
Figure 6-6 Estimated Belt Drive Loss ............ 6-14 Without Tuning Vanes... 2. 2... 6-30 
Figure 6-7 Typical Fan Performance Curves... ...... 6-16 Figure 6-22 System Effect Curves for Various Mitered 
Figure 6-8 System Requirement Curves ........... 6-17 Square Duct Elbows .. 2... ...0.000-. 6-31 
Figure 6-9 Actual Versus Desired Point of Operation .. . . 6-18 Figure 6-23 Non-Uniform Inlet Flows ............ 6-32 
Figure 6-10 | Homologous Performance Curve... ...... 6-19 Figure 6-24 = Non-Uniform Inlet Corrections... 2.2... 6-33 
Figure 6-11 In-Duct Heater 22... ee 6-20 Figure 6-25 System Effect Curves for Inlet Obstructions . . 6-34 
Figure 6-12 Fans: Parallel Operation ............. 6-22 Figure 6-26 System Effect Curves... .....0...0-. 6-35 
Figure 6-13. Fans: Series Operation .............. 6-23 Figure 6-27 System Effect Curves... 0.2. .00.0..00., 6-36 
Figure 6-14 System Effect Factor... 2... 2. eee 6-24 


6-2 Industrial Ventilation 


6.1 INTRODUCTION 


To move air in a ventilation or exhaust system, energy is 
required to overcome the system losses. This energy can be 
in the form of natural convection or buoyancy. Most systems, 
however, require some powered air moving device such as a 
fan or an ejector. 


This chapter will describe the various air moving devices 
that are used in industrial applications, provide guidelines for 
the selection of the air moving device for a given situation, 
and discuss the proper installation of the air moving device in 
the system to achieve desired performance. 


Selection of an air moving device can be a complex task, 
and the specifier is encouraged to take advantage of all 
available information from applicable trade associations as 
well as from individual manufacturers. 


6.2 BASIC DEFINITIONS 


Air moving devices can be divided into two basic classifi- 
cations: ejectors and fans. Ejectors have low operating effi- 
ciencies and are used only for special material handling 
applications. Fans are the primary air moving devices used in 
industrial applications. 


Fans can be divided into three basic groups: axial, centrifu- 
gal, and special types. As a general rule, axial fans are used 
for higher flow rates at lower resistances and centrifugal fans 
are used for lower flow rates at higher resistances. 


6.2.1 Ejectors: (see Figure 6-1)Are used sometimes when 
it is not desirable to have contaminated air pass directly 
through the air moving device. Ejectors are utilized for air 
streams containing corrosive, flammable, explosive, hot, or 
sticky materials that might damage a fan; present a dangerous 
operating situation; or quickly degrade fan performance. 
Ejectors also are used in pneumatic conveying systems. 


6.2.2 Axial Fans: There are three basic types of axial fans: 
propeller, tubeaxial, and vaneaxial (see Figures 6-2 and 6-3). 


Propeller Fans are used for moving air against low static 
pressures and are used commonly for general ventilation. Two 
types of blades are available: disc blade types when there is 
no duct present; narrow or propeller blade types for moving 
air against low resistances (less than I1"wg). Performance is 
very sensitive to added resistance, and a small increase will 
cause a marked reduction in flow rate. 


Tubeaxial Fans (Duct Fans) contain narrow or propeller- 
type blades in a short, cylindrical housing normally without 
any type of straightening vanes. Tubeaxial fans will move air 
against moderate pressures (less than 2"wg). 


Vaneaxial Fans have propeller configuration with a hub 
and airfoil blades mounted in cylindrical housings which 
normally incorporate straightening vanes on the discharge 
side of the impeller. Compared to other axial flow fans, 


vaneaxial fans are more efficient and generally will develop 
higher pressures (up to 8"wg). They are limited normally to 
clean air applications. 


6.2.3 Centrifugal Fans: (see Figures 6-4 and 6-5): These 
fans have three basic impeller designs: forward curved, radial, 
and backward inclined/backward curved. 


Forward curved(commonly called "squirrel cages") impel- 
lers have blades which curve toward the direction of rotation. 
These fans have low space requirements, low tip speeds, and 
are quiet in operation. They usually are used against low to 
moderate static pressures such as those encountered in 
heating and air conditioning work and replacement air 
systems. This type of fan is not recommended for dusts or 
particulates that would adhere to the short curved blades 
and cause unbalance. 


Radial Impellers have blades which are straight or radial 
from the hub. The housings are designed with their inlets and 
outlets sized to produce material conveying velocities. There 
is a variety of impeller types available ranging from "high 
efficiency, minimum material" to "heavy impact resis- 
tance" designs. The radial blade shape will resist material 
buildup. This fan design is used for most exhaust system 
applications when particulates will pass through the fan. 
These fans usually have medium tip speeds and are used 
for a variety of exhaust systems which handle either clean 
or dirty air. 


Backward Inclined/Backward Curved impeller blades are 
inclined opposite to the direction of fan rotation. This type 
usually has higher tip speeds and provides high fan efficiency 
and relatively low noise levels with "non-overloading" horse- 
power characteristics. Inanon-overloading fan, the maximum 
horsepower occurs near the optimum operating point so any 
variation from that point due to a change in system resistance 
will result in a reduction in operating horsepower. The blade 
shape is conducive to material buildup so fans in this group 
should be limited as follows: 


e Single-Thickness Blade: Solid blades allow the unit to 
handle light dust loading or moisture. It should not be 
used with particulates that would build up on the 
underside of the blade surfaces. 


° Airfoil Blade: Airfoil blades offer higher efficiencies 
and lower noise characteristics. Hollow blades erode 
more quickly with material and can fill with liquid in 
high humidity applications. These should be limited to 
clean air service. 


6.2.4 Special Type Fans (see Figure 6-4): [In-line Cen- 
trifugal Fans have backward inclined blades with special 
housings which permit a straight line duct installation. Pres- 
sure versus flow rate versus horsepower performance curves 
are similar to a scroll-type centrifugal fan of the same blade 
type. Space requirements are similar to vaneaxial fans. 


Fans 6-3 


( 
INDUCED AIR 
OL 


~~ PRIMARY AIR 


PRIMARY ALR 


EUECTOR FOR PNEUMATIC CONVITYING 


LD) 


AMERICAN CONFERENCE 
|} OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS = 


6-4 Industrial Ventilation 


IMPELLER v 7 


TUBULAR CENTRIFUGAL FAN-DIRECT DRIVE 


DIFFUSER 


TUBEAXIAL FAN-DIRECT DRIVE 
(IMPELLER DOWNSTREAM) 


GUIDE VANE 


IMPELLER VANEAXIAL FAN-BELT DRIVE 


Reprinted from AMCA Publication 201-90, FANS AND SYSTEMS, by 
permission of the Air Movement and Control Association, Inc.©) 


| TERMINOLOGY FOR AXIAL | 
| AND TUBULAR CENTRIFUGAL } 
FANS 


AMERICAN CONFERENCE 
| OF GOVERNMENTAL : 


Fans 6-5 


HOUSING 


DIVEATER 


BLAST AREA 
DISCHARGE 


OUTLET AREA 


SIDE SHEET 


BACKPLATE 


BEARING 
SUPPORT 


INLET COLLAR 


Reprinted from AMCA Publication 201-90, FANS AND SYSTEMS, by 


permission of the Air Movement and Control Association, Inc.©") 
| OF GOVERNMENTAL _ | CENTRIFUGAL FAN 
| INDUSTRIAL HYGIENISTS | COMPONENTS _ 


AMERICAN CONFERENCE 


6-6 Industrial Ventilation 


Power Exhausters, Power Roof Ventilators are packaged 
units that can be either axial flow or centrifugal type. The 
centrifugal type does not use a scroll housing but discharges 
around the periphery of the ventilator to the atmosphere. 
These units can be obtained with either downward deflecting 
or upblast discharges. 


Fan and Dust Collector Combination: There are several 
designs in which fans and dust collectors are packaged in a 
unit. [f use of such equipment is contemplated, the manufac- 
turer should be consulted for proper application and perform- 
ance characteristics. 


6.3 FAN SELECTION 


Fan selection involves not only finding a fan to match the 
required flow and pressure considerations but all aspects of 


an installation including the air stream characteristics, oper- 
ating temperature, drive arrangement, and mounting. Section 
6.2 discussed the various fan types and why they might be 
selected. This section offers guidelines to fan selection; how- 
ever, the exact performance and operating limitations of a 
particular fan should be obtained from the original equipment 
manufacturer. 


6.3.1 Considerations for Fan Selection: 


CAPACITY 


Flow Rate (Q): Based on system requirements and ex- 
pressed as actual cubic feet per minute (acfm) at the fan inlet. 


Pressure Requirements: Based on system pressure require- 
ments which normally are expressed as Fan Static Pressure 
(FSP) or Fan Total Pressure (FTP) in inches of water gauge 


within the blade passages. For given 


centric housings can also be used as in 


TYPE IMPELLER DESIGN HOUSING DESIGN _ 
7 
; 7 ' Scroll-type, usually designed to permit 
PY re Highest efficiency of all centrifugal fan efficient conversion of velocity pressure 
designs. 9 to 16 blades of airfoil contour to static pressure, thus permitting a high 
= curved away from the direction of rotation. — static efficiency; essential that clearance 
ra Air leaves the impeller at a velocity less and alignment between wheel and inlet 
c than its tip speed and relatively deep bell be very close in order to reach the 
i4 blades provide for efficient expansion maxiumum efficiency capability. Con- 


CENTRIFUGAL FANS 


duty, this will be the highest speed of the 
centrifugal fan designs. 


power roof ventilators, since there is effi- 
cient pressure conversion in the wheel. 


BACKWARD-INCLINED 


BACKWARD-CURVED 


Efficiency is only slightly less than that of 
airfoil fans. Backward-inclined or back- 
ward-curved blades are single thickness. 
9 to 16 blades curved or inclined away 
from the direction of rotation. Efficient for 
the same reasons given for the airfoil fan 
above. 


Utilizes the sarne housing configuration 
as the airfoil design. 


RADIAL 


) 


A 


Simplest of all centrifugal fans and least 
efficient. Has high mechanical strength 
and the wheel is easily repaired. For a 
given point of rating, this fan requires 
medium speed. This classification 
includes radial blades (R) and modifi- 
ed radial blades (M), usually 6 to 10 in 
number. 


FORWARD-CURVED 


) 


Efficiency is less than airfoil and back- 
ward-curved biaded fans. Usually fab- 
ricated of lightweight and low cost con- 
struction. Has 24 to 64 shallow blades 
with both the heel and tip curved forward. 
Air leaves wheel at velocity greater than 
wheel. Tip speed and primary energy 
transferred to the air is by use of high 
velocity in the wheel. For given duty, 
wheel is the smallest of all centrifugal 
types and operates at lowest speed. 


FIGURE 6-4. Types of fans: impeller and housing designs (see facing page) 


Scroll-type, usually the narrowest design 
of all centrifugal fan designs described 

R here because of required high velocity 
discharge. Dimensional requirements of 
this housing are more critical than for air 
foil and backward-inclined blades. 


Scroll is similar to other centrifugal-fan 
designs. The fit between the wheel and 
inlet is not as critical as on airfoil and 
backward-inclinded bladed fans. Uses 
large cut-off sheet in housing. 


at standard conditions (0.075 lbm/ft’). Ifthe required pressure 
is known only at non standard conditions, a density correction 
(see Section 6.3.8) must be made. 


AIR STREAM 


Fans 6-7 


stream). Conform to the standards of the National Board of 
Fire Underwriters, the National Fire Protection Association 
and governmental regulations (see Section 6.3.9). 


Corrosive Applications: May require a protective coating 


or special materials of construction (stainless, fiberglass, etc.) 
Material handled through the fan. When the exhaust air 


contains a small amount of smoke or dust, a backward inclined 
centrifugal or axial fan should be selected. With light dust, 
fume or moisture, a backward inclined or radial centrifugal 
fan would be the preferred selection. If the particulate loading 
is high, or when material is handled, the normal selection 
would be a radial centrifugal fan. 


Elevated Air Stream Temperatures. Maximum operating 
temperature affects strength of materials and therefore must 
be known for selection of correct materials of construction, 
arrangement, and bearing types. 


PHYSICAL LIMITATIONS 


Fan size should be determined by performance require- 
ments. Inlet size and location, fan weight, and ease of main- 


Explosive or Flammable Material: Use spark resistant con- 
struction (explosion-proof motor if the motor is in the air 


PERFORMANCE CURVES PERFORMANCE CHARACTERISTICS* APPLICATIONS 


Highest efficiencies occur 50 to 65% of 
wide open volume. This is also the area of 
good pressure characteristics; the horse- 
power Curve reaches a maximum near the 
peak efficiency area and becomes lower 
toward free delivery, a self-limiting power 
characteristic as shown. 


Generai heating, ventilating and air-con- 
ditioning systems. Used in large sizes for 
clean air industrial applications where 
power savings are significant. 


Operating characteristics of this fan are 
similar to the airfoil fan mentioned above. 
Peak efficiency for this fan is slightly lower 
than the airfoil fan. Normally unstable left of 
peak pressure. 


Same heating, ventilating, and air-con- 
ditioning applications as the airfoil fan. Also 
used in some industrial applications where 
the airfoil blade is not acceptable because 
of corrosive and/or erosion environment. 


SP 10 > Used primarily for material-handing 
2 Ie = applications in industrial plants. Wheel can 
48 2 : dee 
ae ) 6 3 Higher pressure characteristics than the erie yohuky sneer ieee ae 
= Fe above mentioned fans. Power rises con- coated with special material. This desi 
44 tinually to free delivery. pec ial eit 
a la also used for high-pressure industrial 
a7 requirements. Not commonly found in 
t i zi) ites hG icati 
5 4 6 8 id HVAC applications. 
VOLUME FLOW RATE 
Pressure curve is less steep than that of 
7 > backward-curved bladed fans. There is a 
8 Ss dip in the pressure curve left of the peak Used primarily in iow-pressure heating, 
Pt pressure point and highest efficiency ventilating, and air-conditioning applica- 
46 © occurs to the right of peak pressure, 40 to tions such as domestic furnaces, central 
a 50% of wide open volume. Fan should be station units, and packaged air-con- 
- Gi rated to the right of peak pressure. Power ditioning equipment from room air-con- 
42 curve rises continually toward free delivery ditioning units to roof top units. 
5 5 4 G nl - and this must be taken into account when 
VOLUM OW RATE motor is selected. 
- as 5 


ae 


Types of fans: Performance characteristics and applications. (“These performance curves reflect the general characteristics of various fans as commonly 
employed. They are not intended to provide complete selection criteria for application purposes, since other parameters, such as diameter and speed, are 
not defined.) 


6-8 


Industrial Ventilation 


TYPE 


IMPELLER DESIGN 


AXIAL FANS 


SPECIAL DESIGNS 


PROPELLER 


Efficiency is low. Impellers are usually of 
inexpensive construction and limited to 
low pressure applications. Impeller is of 2 
or more blades, usually of single thick- 
ness attached to relatively smal! hub. 
Energy transfer is primarily in form of 
velocity pressure. 


HOUSING DESIGN 


Simple circular ring, orifice plate, or ven- 
turi design. Design can substantially 
influence performance and optimum 
design is reasonably close to the blade 
tips and forms a smooth inlet flow contour 
to the wheel. 


TUBEAXIAL 


Somewhat more efficient than propeller 
fan design and is capable of developing 
a more useful static pressure range. 
Number of blades usually from 4 to 8 and 
hub is usually less than 50% of fan tip 
diameter. Blades can be of airfoil or single 
thickness cross section. 


4- 


Cylindrical tube formed so that the run- 
ning clearance between the wheel tip 
and tube is close. This results in signifi- 
cant improvement over propeller fans. 


VANEAXIAL 


ad 
= 
oO 
> 
— 
= 
Z 
i 
o 


Good design of blades permits medium- 
to high-pressure capability at good effi- 
ciency. The most efficient fans of this type 
have airfoil blades. Blades are fixed or 
adjustable pitch types and hub is usually 
greater than 50% of fan tip diameter. 


This fan usually has a wheel similar to the 
airfoil backward-inclined or backward- 
curved blade as described above. 
(However, this fan wheel type is of lower 
efficiency when used in fan of type.) 
Mixed flow impellers are sometimes 
used. 


POWER ROOF VENTILATORS 


CENTRIFUGAL 


Many models use airfoil or backward- 
inclined impeller designs. These have 
been modified from those mentioned 
above to produce a low-pressure, high- 
volume flow rate characteristic. In addi- 
tion, many special centrifugal impeller 
designs are used, including mixed-flow 
design. 


io | 
i | 
im] 


Cylindrical tube closely fitted to the outer 
diameter of blade tips and fitted with a set 
of guide vanes. Upstream or downstream 
from the impeller, guide vanes convert the 
rotary energy imparted to the air and 
increase pressure and efficiency of fan. 


Cylindrical shell similar to a vaneaxial fan 
housing, except the outer diameter of the 
wheel does not run close to the housing. 
Air is discharged radially from the wheel 
and must change direction by 90 
degrees to flow through the guide vane 
section. 


(sam 


Does not utilize a housing in a normal 
sense since the air is simply discharged 
from the impeller in a 360 degree pattern 
and usually does not include a configura- 
tion to recover the velocity pressure com- 
ponent. 


A great variety of propeller designs are 
employed with the objective of high-vol- 
ume flow rate at low pressure. 


FIGURE 6-4 (continued). Types of fans: impeller and housing design 


Essentially a propeller fan mounted in a 
supporting structure with a cover for 


weather protection and safety considera- 
tions. The air is discharged through the 
annular space around the bottom of the 
weather hood. 


Fans 6-9 


PERFORMANCE CURVES PERFORMANCE CHARACTERISTICS* APPLICATIONS 
x fe 
= 
a. ’ 
+10 o High flow rate but very low-pressure capa- ‘ ’ : 
1g = bilities and maxiumum efficiency is For low-pressure, high-volume air moving 
tu 47 lw reached near free delivery. The discharge applications such as air circulation within a 
ee 46 0 pattern of the air is circular in shape and the space or ventilation through a wail without 
7) 44 air stream swirls because of the action of attached duct work. Used for replacement 
7 uJ the blades and the lack of straightening air applications. 
uJ Z te 
& 40 facilities. 
i0 
a 
4310 
= ie 
© 8b E 
a 8 al ce) iS High flow-rate characteristics with medium- Low- and medium-pressure ducted heat- 
16 48 4 pressure Capabilities. Performance curve ing, ventilating, and air-conditioning 
‘sa q 6 5 includes a dip to the left of peak pressure applications where air distribution on the 
a 4 - ie which should be avoided. The discharge downstream side is not critical. Also used in 
2 = 4 i air pattern is circular and is rotating or whirl- some industrial applications such as dry- 
ae +2 ing because of the propeller rotation and ing ovens, paint spray booths, and fume 
aa ov 10 lack of guide vanes. exhaust systems. 
a 10 
VOLUME FLOW RATE 
h10 
9 BF 40 > High-pressure characteristics with medium General heating, ventilating, and air-con- 
4~ 0° volume flow rate capabilities. Performance ditioning systems in low-, medium-, and 
1 6 78 a curve includes a dip caused by aero- high-pressure applications is of advantage 
td 46 6 dynamic stall to the left of peak pressure, where straight-through flow and compact 
= 4 14 in which should be avoided. Guide vanes cor instalation are required; air distribution on 
a 5 arn rect the circular motion imparted to the air downstream side is good. Also used in 
A 42 by the wheel and improve pressure charac- industrial application similar to the tubeax- 
a O _L 10 teristics and efficiency of the fan. ial fan. Relatively more compact than com- 
a O Zs 4 6 8 10 parable centrifugal-type fans for same duty. 
VOLUME FLOW RATE 
La10 
= 
2 8 10 So Performance is similar to backward-curved 
2 2 fan, except lower capacity and pressure ee : 
Ve 18 & because of the 90 degree change in direc- pals pedal ae nee - 
Wy 6 9 tion of the air flow in the housing. The effi- me : a ti ie le i 
5 A. tes ciency will be lower than the backward- a RIE app he lons. Has straight- 
ey, ao curved fan. Some designs may have a dip rough flow configuration. 
Lo : : 0 in the curve similar to the axial-flow fan. 
ac 
a 0 2 4 6 8 « 10 
VOLUME FLOW RATE 
a 
wi 1O 
= |. For low-pressure exhaust systerns such as 
O Be : ' general factory, kitchen, warehouse, and 
Be idle 70 6 Usually intended to operate without commercial installations where the low- 
| 6F 48 4 attached ductwork and rea to operate pressure rise limitation can be tolerated. 
tu [SP 46 6 against a very low-pressure head. It is usu- Unit is low in first cost and low in operating 
7 4 ' ME Ig & ally intended to have a rather high-volume cost and provides positive exhaust ventila- 
Dob + th flow rate characteristic. Only static pres: tion in the space which is a decided advan- 
n [ ——="" PWR 4 2 sure and static efficiency are shown for this tage over gravity-type exhaust units. The 
a 0 1 ! t ! 30 type of product. centrifugal unit is somewhat quieter than 
a O 2 4 6 8 10 the axial unit desribed below. 
VOLUME FLOW RATE 
tat ; 
Sob 
& 8h +190 > F . For low-pressure exhaust systems such as 
- Be AO) Usually intended to operate without general factory, kitchen, warehouse, and 
1 6F 18. 35 attached ductwork and therefore to operate some commercial installations where the 
bp of Bs FeD against very low-pressure head. Itis usually low-pressure rise limitations can be toler 
is 4 Tye intended to have a high-volume flow rate ated. Unit is low in first cost and low in oper 
A 5 oo characteristic. Only static pressure and ating cost and provides positive exhaust 
4 42 static efficiency are shown for this type of ventilation in the space which is a decided 
a oa d 53 ‘ oo i produci. advantage over gravity-type exhaust units. 
an 2 


VOLUME FLOW RATE 


Types of fans: performance characteristics and applications 


6-10 Industrial Ventilation 


tenance also must be considered. The most efficient fan size 
may not fit the physical space available. 


DRIVE ARRANGEMENTS 


All fans must have some type of power source — usually 
an electric motor. On packaged fans, the motor is furnished 
and mounted by the manufacturer. On larger units, the motor 
is mounted separately and coupled directly to the fan or 
indirectly by a belt drive. A number of standard drive arrange- 
ments are shown in Figures 6-Sa, 6-5b, and 6-5c. 


Direct Drive offers a more compact assembly and assures 
constant fan speed. Fan speeds are limited to available motor 
speeds (except in the case of variable frequency controllers). 
Capacity is set during construction by variations in impeller 
geometry and motor speed. 


Belt Drive offers flexibility in that fan speed can be changed 
by altering the drive ratio. This may be important in some 
applications to provide for changes in system capacity or 
pressure requirements due to changes in process, hood design, 
equipment location or air cleaning equipment. V-belt drives 
must be maintained and have some power losses which can 
be estimated from the chart in Figure 6-6. 


NOISE 


Fan noise is generated by turbulence within the fan housing 
and will vary by fan type, flow rate, pressure, and fan effi- 
ciency. Because each design is different, noise ratings must 
be obtained from the fan manufacturer. Most fans produce a 
"white" noise which is a mixture of all frequencies. In addition 
to white noise, radial blade fans also produce a pure tone at a 
frequency equal to the blade passage frequency (BPF): 


BPF =RPMxNxCF [6.1] 


where: 
BPF = blade passage frequency, Hz 
RPM = rotational rate, rpm 
N = number of blades 
CF = conversion coefficient, 1/60 


This tone can be very noticeable in some installations and 
should be considered in the system design. 


Because of its higher efficiency, the backward inclined 
impeller design is generally the quietest. However, for all fan 
types, non-uniform air flow at the fan inlet or outlet can 
increase the fan noise level. This is another problem related 
to "system effect" (see Section 6.4.1). 


Most fan manufacturers publish sound ratings for their 
products. There are a variety of ways to present the ratings. 
One popular way is to list sound power levels for eight ANSI 
standard octave bands. The sound power levels are typically 
in units called "decibels" (dB). The sound power level is a 
characteristic of a fan that varies with the fan speed and point 
of operation. 


For an installed fan, the surrounding environment affects 
the sound level that is measured or heard. Walls, floors, and 
other equipment reflect and absorb sound to varying degrees. 
The sound that reaches the listener will be different than the 
fan’s rated sound power level. Typical sound measuring de- 
vices detect sound with a microphone and display sound 
pressure level in decibels. This sound pressure is an environ- 
ment-dependent measurement that changes with listener lo- 
cation and/or environment changes. 


While the decibel unit is used for sound power and sound 
pressure, the two measures are not interchangeable. For in- 
stance, 70 dB sound power is not 70 dB sound pressure. The 
decibel is not an absolute unit of measure. It is a ratio between 
a measured quantity and an agreed reference level. Both dB 
scales are logarithmic. The sound power is the log of the ratio 
of two power levels. The sound pressure is the log of the ratio 
of two pressure levels. The sound power scale uses a reference 
of 10-!? watts. The sound pressure scale uses a reference of 20 
x 10° N/M?. 


For an installed fan, the sound pressure levels are usually 
measured in dB using the "A" weighting scale. The A-weight- 
ing is used to measure environmental noise as it most closely 
reflects the human auditory response to noise of various 
frequencies. A sound level meter set on the "A" scale auto- 
matically integrates the noise ofall frequencies to give asingle 
dBA noise measurement. Expanded detail can be obtained by 
taking noise measurements with a meter capable of measuring 
the sound pressure level in each octave band. Such detail can 
help indicate the predominant source of a noise. 


The topic of sound is quite broad and there are many 
reference texts available to cover it. For a concise introduction, 
the ASHRAE Fundamentals Handbook” is a good starting 
point. 


SAFETY AND ACCESSORIES 


Safety Guards are required. Consider all danger points such 
as inlet, outlet, shaft, drive and cleanout doors. Construction 
should comply with applicable governmental safety require- 
ments, and attachment must be secure. 


Accessories can help in the installation and in future main- 
tenance requirements. Examples might include drains, 
cleanout doors, split housings, and shaft seals. 


FLOW CONTROL 


There are various accessories that can be used to change 
fan performance. Such changes may be required on systems 
that vary throughout the day or for reduction in flow rate in 
anticipation of some future requirement. Dampers, variable 
pitch blades, and speed control are three common accessories 
used with fans. 


Dampers are installed directly on the fan inlet or outlet. 
Because they are in the air stream, dampers can build up with 


Fans 6-11 


SW - Single Width DW - Double Width 
SI - Single Iniet Di - Double Iniet 


Arrangements 1. 3. 7 and 8 are also available with bearings mounted 
on pedestals or base set independent of the fan housing 


ARR. 1 SWSi_ For belt drive or di- 
rect connection. Impeller overhung 
Two bearings on base 


ARR. 2 SWS! For belt drive or di- ARR. 3 SWSI For belt drive or di- ARR. 3 DWDI For belt drive or di- 
rect connection. impeller overhung. rect connection. One bearing on rect connection One bearing on 
Bearings in bracket supported by each side and supported by fan each side and supported by fan 
fan housing. housing. housing 


ARR. 4 SWS! For direct drive. Im- ARR. 7 SWSI For belt drive or di- ARR. 7 DWDI For belt drive or di- 
peller overhung on prime mover rect connection. Arrangement 3 rect connection. Arrangement 3 
shaft. No bearings on fan. Prime plus base for prime mover pilus base for prime mover 

mover base mounted or integrally 

directly connected. 


ARR. 8 SWS! For belt drive or di- ARR. 9 SWS] For belt drive Im- ARR. 10 SWSI For belt drive Im- 
rect. connection. Arrangement 1 peller overhung, two bearings. with pelier overhung, two bearings, with 
plus extended base for prime prime mover outside base prime mover inside base 

mover. 


Reprinted from AMCA Publication 99-86, STANDARDS HAND- 
rae by permission of the Air Movement and Control Association, 
Inc.” 


DRIVE ARRANGEMENTS 
FOR CENTRIFUGAL 
FANS 


AMERICAN CONFERENCE 
» OF GOVERNMENTAL 
|} INDUSTRIAL HYGIENISTS 


6-12 


industrial Ventilation 


SW - Single Width DW - Doubie Width 
SI - Single Inlet Di - Double Intet 


ARR.1SWSI WITH INLET BOX For 
belt drive or direct connection. Impel- 
ler overhung, two bearings on base. 
Inlet box may be self-supporting. 


ARR. 3 SWSI WITH INDEPENDENT 
PEDESTAL For belt drive or direct, 
connection fan. Housing is self-sup- 
porting. One bearing on each side 
supported by independent pedestals. 


ARR. 3 SWS! WITH INLET BOX AND 
INDEPENDENT PEDESTALS For 
belt drive or direct connection fan. 
Housing is self-supporting. One 
bearing on each side supported by in- 
dependent pedestals with shaft ex- 
tending through inlet box. 


ARR. 3 OWDI WITH INDEPENDENT ARR.3 DWDI WITH INLET BOX AND 


PEDESTAL For belt drive or direct 
connection fan. Housing is self-sup- 


INDEPENDENT PEDESTALS For 
belt drive or direct connection fan. 


ARR. 8 SWS! WITH INLET BOX For 
belt drive or direct connection. |mpel- 
ler overhung, two bearings on base 


porting. One bearing on each side 
supported by independent pedestals. 


Housing is’ self-supporting. One 
bearing on each side supported by in- 
dependent pedestals with shaft ex- 
tending through inlet box. 


plus extended base for prime mover. 
Intet box may be self-supporting. 


Reprinted from AMCA Publication 99-86, STANDARDS HAND- 
BOR by permission of the Air Movement and Control Association, 
inc.’ 


DRIVE ARRANGEMENTS 
FOR CENTRIFUGAL 
FANS 


AMERICAN CONFERENCE 
| OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


Fans 6-13 


. aaa okt Inlet box and 
\ Inlet 4 ie evase are 


ARR. 1 TWO STAGE 
For belt drive or direct connection. Impeller overhung. Two bearings 
located either upstream or downstream of impeller. 


N 


ARR. 3 ARR. 4 4 TWO STAGE 


For belt drive or direct For direct connection. impeller 
connection. Impeller between overhung on motor shaft. No 
bearings that are on internal bearings on fen. Motor on 
supports. Drive through inlet. internal supports. 


-afe sl 
| 


Peel. | 


ARR, 7 ARR. 8 (1 of 2 stage) 
For belt drive or direct connection. For belt drive or direct 


Arr. 3 plus common base for prime connection. Arr. 1 plus 
mover, 


comrnon base for prime mover. 


c: 


Motor on Casing ARR. G Motor on Integral Base 
For belt drive. Impeller overhung. Two bearings on internal supoorts. 
Motor on casing or on integral base. Drive through belt fairing. 

NOTE: Ali fan orientations may be horizontal or vertical. 


Reprinted from AMCA Publication 99-86 Standards Handbook, 
by permission of the Air Movement and Control Association Inc. (6.1) 


AMERICAN CONFERENCE | DRIVE ARRANGEMENTS FOR 
 CAVEDNMPT A) | AXIAL FANS WITH OR WITHOUT 
OF GOVERNMENTAL — | 


a , ! EVASE AND INLET BOX 
| INDUSTRIAL HYGIENISTS fe—47—gg_ rome oY 


6-14 Industrial Ventilation 


2 3 4 6 810 20 30 40 60 80100 200 300400 600 


Ss Sa ee ee ee ee a a 
60 PAR eiRe eee Ries 
i eters 

SS a Ste liseskieeiesS 

a: 

2 45 wd 

2 

5 10 | 

5 8 | 

= 

B 

S 

ut 

> 

Cc 

Q 


a8 
= 
a 
af: 
= 
if 
i 


is Sa, 
are ee 


ale et ene TEN cd al 
oo 


CT ATT 


ela ew oe 
2 KS Sa VO a) Fe eG 


° 
w 
° 
> 
o 
fon) 
oO 
ror) 
ar 


MOTOR POWER OUTPUT, hp 
HIGHER BELT SPEEDS TEND TO HAVE HIGHER LOSSES 
THAN LOWER BELT SPEEDS AT THE SAME HORSEPOWER 
*Drive losses are based on the conventional V-belt which has been the “work horse” of the drive industry 


for several decades. 


EXAMPLE 


@ Motor power output, H,,., is determined to be 13.3 hp 
@ The belts are the standard type and just warm to the touch immediately after shutdown 
@ From chart, drive loss = 5.1% 

@ Drive loss, H, 0.051 x 13.3 


@ Fan power input, H 


not it nog 


Reprinted from AMCA Publication 203-90, FIELD PERFORMANCE 
MEASUREMENT OF FAN SYSTEMS, by permission of the Air 
Movement and Control Association, Inc." 


| AMERICAN CONFERENCE | ESTIMATED BELT 
| OF GOVERNMENTAL 


_DRIVE LOSS _ 
eee 


material and may not be acceptable on material handling fans. 
Two types of dampers are available: 


¢ Outlet Dampers mount on the fan outlet to add 
resistance to the system when partially closed. These are 
available with both parallel and opposed blades. Selection 
depends on the degree of control required (opposed blade 
dampers will control the flow more evenly throughout the 
entire range from wide open to closed). 


¢ Jnlet Dampers mount on the fan inlet to pre-spin air into 
the impeller. This reduces fan output and lowers 
operating horsepower. Because of the power savings, 
inlet dampers should be considered when the fan will 
operate for long periods at reduced capacities. 


Variable pitch blades are available with some axial-type 
fans. The fan impellers are designed to allow manual or 
automatic changes to the blade pitch. "Adjustable" impellers 
have a blade pitch that can be manually changed when the fan 
is not running. "Variable" impellers include devices to allow 
the blade pitch to be changed pneumatically or hydraulically 
while the fan is operating. 


A Variable Frequency Drive (VFD) may also be used to 
control flow. A VFD will control the fan speed, rather than 
varying the fan inlet flow conditions or the outlet area to 
change the fan’s point of operation. This type of control varies 
both the flow rate and the fan static pressure. 


The VFD contro! unit is connected in-line between the 
electric power source and the fan motor. It is used to vary the 
voltage and frequency of the power input to the motor. The 
motor speed will vary linearly with the line frequency. Most 
VFD applications use a direct drive arrangement; however, 
belt drives are occasionally used. 


For atypical system with fixed physical characteristics, the 
attainable points of operation will fall on the system curve. 
For example, Figure 6-10 shows points Al and A2 onasystem 
curve. These two points of operation can be attained with a 
VFD by adjusting it for speeds of RPM, or RPM). This will 
result in fan curve PQ, or PQs, respectively. 


VFDs do have disadvantages. They may have a low speed 
limitation. Most AC motors are designed to operate at their 
nameplate speeds. If a VFD is used to run a motor well below 
its nominal speed, the motor’s efficiency will be reduced and 
losses will increase. This can increase motor heating and may 
cause damage. 


The VFD can cause harmonic distortion in the electrical 
input lines from the power source. This may affect other 
electrical equipment on the same power system. Such distor- 
tion can be reduced with the addition of isolation transformers 
or line inductors. 


To properly apply a VFD, the equipment supplier needs to 
know about its intended usage, about the building’s power 
supply and about other electrical equipment in use. In general, 


Fans 6-15 


for applications where the minimum system air flow is 80% 
or more of the maximum system air flow, the VFD’s losses 
and higher initial cost may make use of the inlet damper a 
better choice for flow control. 


An advantage of the VFD or the Variable Pitch Blade over 
the dampers is often a dramatic power and noise reduction. 
However, these accessories usually require additional con- 
trolling equipment. An advantage of dampers is their rela- 
tively simple installation and use and their lower initial costs. 


6.3.2 Rating Tables: Fan size and operating RPM and 
Power usually are obtained from a rating table based on 
required air flow and pressure. Tables are based on FTP or FSP: 


Fan TP =(SPoutiet + VPoutiet) — (SPintet + VPiniet) [6.2] 


Fan SP = SPoutiet — SPintet — VPintet [6.3] 


Fan Rating Tables are based on requirements for air at stand- 
ard conditions (0.075 lbm/ft*). If other than standard condi- 
tions exist, the actual pressure must be converted to standard 
conditions. See Section 6.3.8, "Selection at Air Densities 
Other Than Standard." 


The most common form of table is a "multi-rating table" 
(see Table 6-1) which shows a range of capacities for a 
particular fan size. For a given pressure, the highest mechani- 
cal efficiency usually will be in the middle third of the "CFM" 
column. Some manufacturers show the rating of maximum 
efficiency for each pressure by underscoring or similar indi- 
cator. In the absence of such a guide, the design engineer must 
calculate the efficiency from the efficiency equation 


QxFTP — Qx(FSP+VPyutet) 


n= [6.4] 
CFxPWR CFxPWR 


where: 
n= Mechanical efficiency 
Q= Volumetric flow rate, cfm 
FTP = Fan total pressure,"wg 
FSP = Fan Static Pressure,"wg 
PWR= Power requirement, hp 
CF = Conversion Coefficient, 6362 


Even with a multi-rating table, it is usually necessary to 
interpolate in order to select fan RPM and BHP for the exact 
conditions desired. In many cases a double interpolation will 
be necessary. Straight line interpolations throughout the 
multi-rating table will introduce negligible errors. 


Certain types of fans may be offered in various Air Move- 
ment and Control Association©» performance classes identi- 
fied as | through IV. A fan designated as meeting the 
requirements of a particular class must be physically capable 
of operating at any point within the performance limits for 
that class. Performance limits for each class are established in 
terms of outlet velocity and static pressure. Multi-rating tables 


6-16 industrial Ventilation 


TABLE 6-1. Exampie of Multi-Rating Table 


Inlet diameter: 13" 0.D. 
Outlet area: .930 sq. ft. inside 


Wheel diameter: 2254" 
Wheel circumference: 5.92 ft. 


18"SP fa 20"SP 22"SP 


2°SP 4"SP 12"SP 14"SP 16"SP 
CFM | OV 
PM BHP|RPM BHP BHP BHP|RPM BHP|RPM BHP|RPM BHP|RPM BHP |RPM BHP 
930 | 1000| 843 0.57 | 1176 1.21 | 1434 1.93 2.75 4.59 | 2184 5.62 | 2333. 6.68 | 2475 7.81 | 2610 9.01 } 2738 = 10.2 
1116 | 1200} 853 0.67 | 1183 1.35 | 1439 2.12 2.98 489 | 2182 5.95 | 2333 7.07 | 2473 8.23 | 2606 9.45 | 2733 10.7 
1302 ; 1400] 866 0.77; 1191 1.51 | 1445 2.33 3.22 5.23 | 2183 6.31 | 2333. 7.47 | 2474 «8.68 | 2606 9.95 | 2731 11.2 
1488 | 1600) 882 0.89 | 1201 1.69 | 1453 2.56 3.50 5.59 | 2188 6.72 | 2337. 7.92 | 2474 9.13 | 2606 104 | 2734 11.8 
1674 | 1800} 899 1.01 {1213 1.88 | 1463 281 3.81 5.98 } 2194 7.16 } 2340 8.38 | 2479 9.67 | 2610 11.0 | 2735 12.4 
1860 | 2000} 917 1.14) 1227 2.09 } 1474 3.09 4.13 6.39 | 2199 7.62 | 2344 8.89 | 2484 10.2 | 2613 11.6 | 2735 = 13.0 
2046 | 2200} 937 1.29 | 1242 2.32 | 1484 3.37 4.48 6.84 | 2206 8.13 | 2351 9.43 | 2487. 10.8 | 2618 122] 2741 13.6 
2232 | 2400) 961 1.45] 1257 2.56 | 1497 3.68 4.85 7.33 | 2212 8.64 | 2357 10.0 | 2493, 11.4 | 2622) «128 } 2745 14.3 
2418 } 2600} 984 1.62} 1275 281 } 1513 4.02 5.25 7.84 | 2222 9.22 | 2364 «10.6 } 2501 12.1 | 2631 136] 2750 15.1 
2790 } 3000 | 1038 2.02 | 1313. 3.36 | 1543 4.73 6.11 8.96 | 2241 10.4 | 2383 12.0 | 2517) 13.5 | 2644 «15.1 1 2766 «16.7 
3162 | 3400] 1099 2.50 | 1358 3.99 | 1580 5.52 7.05 10.2 | 2265 11.8] 2405 13.4 | 2538 15.1 | 2665 16.8 | 2783 18.5 
3534 | 3800) 1164 3.07 | 1407 4.69 | 1620 6.37 8.09 11.5 | 2290 13.3 | 2428 15.0 | 2562 16.8 | 2684 =18.6 | 2803 20.5 
3906 | 4200 | 1232 3.75 | 1462 5.48 | 1665 7.31 9.19 12.9 | 2320 14.8 | 2458 16.8 |] 2587 18.7 } 2708 20.6 | 2825 22.5 
4278 | 4600 | 1306 4.56 | 1520 6.39 {1717 8.38 10.4 14.5 | 2355 16.5 | 2489 186} 2614 20.6] 2736 22.7 | 2852 24.8 
4650 | 5000 | 1380 5.49 | 1582 7.41 770 9.53 117 16.1 | 2390 183] 2521 20.5 | 2645 22.7 | 2766 25.0 | 2883 27.3 
5022 | 5400] 1457 6.56 | 1647 8.57 | 1827 108 13.1 17.8 | 2428 20.2 | 2558 = =22.6 | 2681 =25.0 | 2798 = 27.3 
5394 [5300 1535 7.79 | 1719 9.93 | 1885 12.2 147 19.7 | 2469 22.2 | 2594 24.7 | 277 27.3 { 2830 29.8 


Performance shown is for fans with outlet ducts and with inlet ducts. BHP shown does not include belt drive losses. 


usually will be shaded to indicate the selection zones for 
various classes or will state the maximum operating RPM. 
This can be useful in selecting equipment, but class definition 
is only based on performance and will not indicate quality of 
construction. 


Capacity tables which attempt to show the ratings for a 
whole series of homologous fans on one sheet cannot be used 
accurately unless the desired rating happens to be listed on the 
chart. Interpolation is practically impossible since usually 
only one point of the fan curve for a given speed is defined in 
such a table. 


f 
p ye ame FOS 
(PWR) 
aes 
=e 
ud a 
Oc eed 
oy 
o ts J a ee \ 
Luo 
“% oO a 
a a ge 


FLOW RATE (Q) 


FIGURE 6-7. Typical fan performance curve 


Today, most fan manufacturers have "electronic catalogs" 
available. These catalogs are computer programs which can 
be used to calculate the correct fan speed and horsepower 
based on input data such as desired flow rate and fan static 
pressure or fan total pressure. Some electronic catalogs in- 
clude estimates of the affects of various fan accessories such 
as dampers and inlet boxes. 


6.3.3 Point of Operation: Fans are usually selected for 
operation at some fixed condition or single "Point of Opera- 
tion." Both the fan and the system have variable performance 
characteristics which can be represented graphically as curves 
depicting an array of operating points. The actual "point of 
operation" will be the one single point at the intersection of 
the fan curve and the system curve. 


Fan Performance Curves: Certain fan performance vari- 
ables are usually related to volumetric flow rate in graphic 
form to represent a fan performance curve. Figure 6-7 is a 
typical representation where Pressure (P) and power require- 
ment (PWR) are plotted against flow rate (Q). Other variables 
also may be included and more detailed curves representing 
various fan designs are provided in Figure 6-4. Pressure can 
be either FSP or FTP. This depends on the manufacturer’s 
method of rating. 


It should be noted that a fan performance curve is always 
specific to a fan of given size operating at a single rotation 
rate (RPM). Even with size and rotation rate fixed, it should 
be obvious that pressure and power requirements vary over a 
range of flow rates. 


System Requirement Curves: The duct system pressure also 
varies with volumetric flow rate. Figure 6-8 illustrates the 
variation of pressure (P) with flow rate (Q) for three different 
situations. The turbulent flow condition is representative of 


FLOW RATE (Q) 


TURBULENT FLOW 
AP = CO? 


“FLOW RATE (Q)_ 
LAMINAR FLOW 
AP =: CO 


FLOW RATE (Q) 


CONSTANT HEAD 
AP=C 


FIGURE 6-8. System requirement curves 


duct losses and is most common. In this case, the pressure loss 
varies as the square of the flow rate. The laminar flow condi- 
tion is representative of the flow through low velocity filter 
media. Some wet collector designs operate at or close to a 
constant loss situation. 


The overall system curve results from the combined effects 
of the individual components. 


6.3.4 Matching Fan Performance and System 
Requirement: A desired point of operation results from the 
process of designing a duct system and selecting a fan. Con- 


Fans 6-17 


sidering the system requirement or fan performance curves 
individually, this desired point of operation has no special 
status relative to any other point of operation on the individual 
curve. Figure 6-9 depicts the four general conditions which 
can result from the system design fan selection process. 


There are a number of reasons why the system design, fan 
selection, fabrication, and installation process can result in 
operation at some point other than design. When this occurs, 
it may become necessary to alter the system physically which 
will change the system requirement curve and/or cause a 
change in the fan performance curve. Because the fan per- 
formance curve is not only peculiar to a given fan but specific 
to a given rotation rate (RPM), a change of rotation rate can 
be relatively simple if a belt drive arrangement has been used. 
The "Fan Laws" are useful when changes of fan performance 
are required. 


6.3.5 Fan Laws: Fan laws relate the performance vari- 
ables for any homologous series of fans. A homologous series 
represents a range of sizes where all dimensional variables 
between sizes are proportional. The performance variables 
involved are fan size (SIZE), rotation rate (RPM), gas density 
(p), flow rate (Q), pressure (P), power requirement (PWR), 
and efficiency (n). Pressure (P) may be represented by total 
pressure (TP), static pressure (SP), velocity pressure (VP), fan 
static pressure (FSP), or fan total pressure (FTP). 


At the same relative point of operation on any two perform- 
ance curves in this homologous series, the efficiencies will be 
equal. The fan laws are mathematical expressions of these 
facts and establish the inter-relationship of the other variables. 
They predict the effect of changing size, speed, or gas density 
on capacity, pressure, and power requirement as follows: 


3 
SIZE, \'(RPM 
Q,=aQ ca | pe [6.5] 
; (32) (Fae | 
SIZE, (RPM, \) 
seiieaiie] 
SIZE, } [RPM, | | p, 
5 3 
PWR, = Pwr, SIZE2) (RPM ) (02 67 
SIZE, | | RPM, } | p, 


As these expressions involve ratios of the variables, any 
convenient units may be employed so long as they are consis- 
tent. Size may be represented by any linear dimension since 
all must be proportional in homologous series. However, 
impeller diameter is the most commonly used dimension. 


6.3.6 The Effect of Changing Rotation Rate or Gas 
Density: In practice, these principles are normally applied to 
determine the effect of changing only one variable. Most often 
the fan laws are applied to a given fan size and may be 
expressed in the simplified versions which follow: 


° For changes of rotation rate: 


6-18 Industrial Ventilation 


DESIRED 


ACTUAL 


FLOW RATE (Q) FLOW RATE 
FAN AND SYSTEM MATCHED B. WRONG FAN. 


DESIRED - 7 
“\ \ . 
» \ & 
\ a 


ACTUAL —~ eo 


ra 


ae Sle 


SSURE (P) 


PRE 


Bore 


LOW RATE (Q) FLOW RATE (Q) 
WRONG SYSTEM. . BOTH FAN AND SYSTEM WRONG 


AMERICAN C LONF ERENCE. |. .ACTOAL. VERSUS "DESIRED 
OF GOVERNMENTAL asap OF O oe TION 


ee 


Flow varies directly with rotation rate; pressure var- 
ies as the square of the rotation rate; and power varies 
as the cube of the rotation rate: 


RPM 
Q, =Q 2] 6.8 
: (eM 6. 
2 
RPM 
P, =P. 7 6.9 
2 (FM [ ] 
RPM, )° 
PWR, = PWR, 2 [6.10] 
RPM, 


¢ For changes of gas density: 


Flow is not affected by a change in density; pressure 
and power vary directly with density: 


Q2=Q, [6.14] 
P, = P(e) [6.12] 
Py 
PWR, = pwr, 2) 16.13] 
: 


6.3.7 Limitations on the Use of Fan Laws: These ex- 
pressions are equations which rely on the fact that the per- 
formance curves are homologous and that the ratios are for 
the same relative points of rating on each curve. Care must 
be exercised to apply the laws between the same relative 
points of rating. 


Figure 6-10 contains a typical representation of two ho- 
mologous fan performance curves, PQ; and PQ). These could 


(OS Se ee 


80 is a ~4I ay 


FLOW RATE {Q) 


FIGURE 6-10. Homologous performance curves 


Fans 6-19 


be the performances resulting from two different rotation 
rates, RPM, and RPM). Assuming a point of rating indicated 
as A; on PQ, there is only one location on PQ) with the same 
relative point of rating and that is at Ay. The A, and A; points 
of rating are related by the expression 


2 
Q 
Pay = | a [6.14] 
4 


The equation can be used to identify every other point that 
would have the same relative point of rating as A, and Ao. 
The line passing through "A>, A," and the origin locates all 
conditions with the same relative points of rating. These lines 
are more often called "system lines" or "system curves." As 
discussed in Section 6.3.3, there are a number of exceptions 
to the condition where system pressure varies as the square of 
flow rate. These lines representing the same relative points of 
rating are "system lines" or "system curves" for turbulent flow 
conditions only. 


Where turbulent flow conditions apply, it must be under- 
stood that the system curves or lines of relative points of rating 
represent a system having fixed physical characteristics. For 
example the, "Bz - B;" line defines another system which has 
lower resistance to flow than the "Aj - A," system. 


Special care must be exercised when applying the fan laws 
in the following cases: 


1. Where any component of the system does not follow 
the "pressure varies as the square of the flow rate” rule. 


2. Where the system has been physically altered. or for 
any other reason operates on a different system line. 


6.3.8 Fan Selection at Air Density Other Than 
Standard: As discussed in Section 6.3.6, fan performance is 
affected by changes in gas density. Variations in density due 
to normal fluctuations of ambient pressure, temperature, and 
humidity are smal! and need not be considered. Where tem- 
perature, humidity, elevation, pressure, gas composition or a 
combination of two or more cause density to vary by more 
than 5% from the standard 0.075 lbm/ft*, corrections should 
be employed. 


Rating tables and performance curves as published by fan 
manufacturers are based on standard air. Performance vari- 
ables are always related to conditions at the fan inlet. Fan 
characteristics are such that volumetric flow rate (Q) is unaf- 
fected but pressure (P) and power (PWR) vary directly with 
changes in gas density. Therefore, the selection process re- 
quires that rating tables are entered with actual volumetric 
flow rate but with a corrected or equivalent pressure. 


The equivalent pressure is that pressure corresponding to 
standard density and is determined from Equation 6.12 as 
follows: 


6-20 Industrial Ventilation 


where: 
P, = Equivalent Pressure 
P, = Actual Pressure 
p,= Actual density, lbm/ft 


The pressures (P, and P,) can be either Fan Static Pressure 
or Fan Total Pressure in order to conform with the manufac- 
turer’s rating method. 


The fan selected in this manner is to be operated at the 
rotation rate indicated in the rating table and actual volumetric 
flow rate is that indicated by the table. However, the pressure 
developed is not that indicated in the table but is the actual 
value. Likewise, the power requirement is not that of the table 
as it also varies directly with density. The actual power 
requirement can be determined from Equation 6.13 as fol- 
lows: 


PWR, = PWR,( 5 Pe =) 


where: 

PWR, = Actual Power Requirement 

PWR, = Power Requirement in Rating Table 

a= Actual Density, Ibm/ft® 

Fan selection at non-standard density requires knowledge of 
the actual volumetric flow rate at the fan inlet, the actual 
pressure requirement (either FSP or FTP, depending on the 
rating table used) and the density of the gas at the fan inlet. 
The determination of these variables requires that the system 
design procedure consider the effect of density as discussed 
in Chapter 5. 


EXAMPLE 


Consider the system illustrated in Figure 6-11 where the 
heater causes a change in volumetric flow rate and density. 
For simplicity, assume the heater has no resistance to flow 
and that the sum of friction losses will equal FSP. Using the 
Multi-Rating Table, Table 6-1, select the rotation rate and 
determine power requirements for the optional fan locations 
ahead of or behind the heater. 


Location 1: Fan ahead of the heater (side "A" to "B" in Figure 
6-11). 
Step 1. Determine actual FSP 
FSP = 1"wg+ 3 "wg 
= 4"wg at 0.075 Ibm/ft®. 
Step 2a. Density at fan inlet is standard. Therefore, 


enter rating table with actual volumetric flow 
rate at fan inlet, 1000 acfm, and FSP of 4 "wg. 


b. Interpolation from Table 6-1 results in: 
RPM = 1182 rpm 
PWR = 1.32 bhp 


Step 3. The fan should be operated at 1182 rpm and 


actual power requirement will be 1.32 bhp. 
Location 2: Fan behind the heater (side "B" to "C" in Figure 
6-11). 
Step 1. Determine actual FSP 
FSP = 1"wg + 3 "wg (as in explanation) 
= 4"wg at 0.0375 Ibm/ft® 
Step 2a. Density at fan inlet is not standard and a pres- 


sure correction must be made (using Equation 
6.12) to determine equivalent FSP. 


1 | 

| PH 
on ™ i i : E m a ee ote ee ee 

ee A a 
a hailey nt 1 E pric on a phases sind ase il i ip mateo anal 
R 
| 
(A) (B) cc 
eed Sa i 
bh 1000 ACFM — 2000 ACFM ——————-- — 
| 
ee 70 F ee 600 Pe 
| 3 3 
b=—————— 0.075 LBS/FT _— — 0.0375 LBS/FT. ——————- 
[ee 1 "wg FRICTION LOSS @ 70 F ef “wq FRICTION LOSS @ 600 F ~~ Se 
(given) (given) 


FIGURE 6-11. In-duct heater 


FSP, = rep { 2078 = | eels } =8"wg 
Pa 0.0375 


Now, enter rating table with actual volumetric 
flow rate at fan inlet, 2000 acfm, and equiva- 
lent FSP, 8"wg. 


b. Interpolation from Table 6-1 results in: 
RPM = 1692 rpm 
PWR = 4.39 bhp 


Step 3a. The fan should be operated at 1692 rpm, but 
actual power requirements will be affected by 
the density and can be determined by using 
Equation 6.13. 


PWR, = PWR; Me 5) 420( S83) 


= 2.2bhp 


b. It should also be noted that a measurement of 
FSP will result in the value of 4"wg (actual) 
and not the equivalent value of 8"wg. 


It will be noted that, regardless of location, the fan will 
handle the same mass flow rate. Also, the actual resistance to 
flow is not affected by fan location. It may appear then that 
there is an error responsible for the differing power require- 
ments of 1.32 bhp versus 2.2 bhp. In fact, the fan must work 
harder at the lower density to move the same mass flow rate. 
This additional work results in a higher temperature rise in the 
air from fan inlet to outlet. A fan located ahead of the heater 
will require less power and may be quieter due to the lower 
rotational speed. 


6.3.9 Explosive or Flammable Materials: When convey- 
ing explosive or flammable materials, it is important to rec- 
ognize the potential for ignition of the gas stream. This may 
be from airborne material striking the impeller or by the 
physical movement of the impeller into the fan casing. 
AMCA® and other associations offer guidelines for both the 
manufacturer and the user on ways to minimize this danger. 
These involve more permanent attachment of the impeller to 
the shaft and bearings and the use of buffer plates or spark- 
resistant alloy construction. Because no single type of con- 
struction fits all applications, it is imperative that both the 
manufacturer and the user are aware of the dangers involved 
and agree on the type of construction and degree of protection 
that is being proposed. 


NOTE: or many years, aluminum alloy impellers have 
been specified to minimize sparking if the impeller were 
to contact other steel parts. This is still accepted, but 
tests by the U. S. Bureau of Mines©® and others have 
demonstrated that impact of aluminum with rusty steel 


Fans 6-21 


creates a"'Thermite" reaction and thus possible ignition 
hazards. Special care must be taken when aluminum 
alloys are used in the presence of steel. 


6.4 FAN INSTALLATION AND MAINTENANCE 


Fan rating tests for flow rate, static pressure, and power 
requirements are conducted under ideal conditions which 
include uniform straight air flow at the fan inlet and outlet. 
However, if in practice duct connections to the fan cause 
non-uniform air flow, fan performance and operating effi- 
ciency will be affected. Location and installation of the fan 
must consider the location of these duct components to mini- 
mize losses. If adverse connections must be used, appropriate 
compensation must be made in the system calculations. Once 
the system is installed and operating, routine inspection and 
maintenance will be required if the system is to continue to 
operate at original design levels. 


6.4.1 System Effect: System effect is defined as the esti- 
mated loss in fan performance from this non-uniform air flow. 
Figure 6-14 illustrates deficient fan system performance. The 
system pressure losses have been determined accurately and 
a suitable fan selected for operation at Point |. However, no 
allowance has been made for the effect of the system connec- 
tions on fan performance. The point of intersection between 
the resulting fan performance curve and the actual system 
curve is Point 3. The resulting flow rate will, therefore, be 
deficient by the difference from 1 to 3. To compensate for this 
system effect, it will be necessary to add a "system effect 
coefficient" to the calculated system pressure. This will be 
equal to the pressure difference between Points | and 2 and 
will have to be added to the calculated system pressure losses. 
The fan then will be selected for this higher pressure (Point 
2) but will operate at Point I due to loss in performance from 
system effects. 


One commonly neglected system effect is a duct elbow at the 
fan inlet. For example, consider the fan shown in Figure 6-16. 


This fan has a four-piece 90° round duct elbow immediately 
in front of the inlet. There are no turning vanes inside the duct. 
The required flow rate is 5000 cfm and the system pressure 
losses are 8"wg at standard conditions (0.075 Ib/ft3). Select- 
ing a fan without the system effect, using Table 6-1, would 
result in a fan speed of 1987 rpm and power consumption of 
13.02 hp. 


With the elbow at the inlet, the air flow into the fan inlet 
will be degraded. Such a change in the air flow requires use 
of a system effect coefficient to select a fan that overcomes 
the degradation in performance. The system effect coefficient 
is used to determine a correction value, in inches water gauge, 
to be added to the system pressure losses. 


In this example, the duct diameter is 24" with a turning 
radius of 48". This is a radius-to-diameter (r/d) ratio of 2.0. In 
Figure 6-21, Item C, we find the system effect curve to use is 


6-22 industrial Ventilation 


FLOW RATE. 


TWO IDENTICAL FANS TWO DIFFERENT FANS 
RECOMMENDED SATISFACTL IRY 


lOTES: 

1. TO ESTABLISH COMBINED 
COMBINED AIR FLOW 
OF INDIVIDUAL FAN Al 


; bas - SU 
LOW RATES AT 


RA 
R 
PONTS OF EQUAL PRESSURE 


TO ESTABLISH SYSTEM CURVE, INCLUDE 
LOSSES IN INDIVIDUAL FAN CONNECTIONS. 


SYSTEM CURVE MUST INT ERSECT 
FAN CURVE OR HIGHER PRES 
MAY HANDLE MORE AiR ALONE. 


TWO DIFFERENT FANS 
UNSATISFACTORY 


WHEN SYSTEM CURVE DOES NOT CROSS COMBINED FAN 
CURVE, OR CROSSES PROJECTED COMBINED CURVE 
BEFORE FAN B, FAN B WILL HANDLE MORE AIR THAN 
FANS A AND BIN PARALLEL. 


AMERICAN C JNE ERENCE | FANS 
OF GOVERNMENTAL i PARALLEL OPERA TION 


INDUSTRIAL: HYGIENISTS | DATE TQ 9G *YricuRE 


Fans 6-23 


TWO IDENTICAL FANS TWO DIFFERENT FANS 
RECOMMENDED FOR BEST EFFICIENCY SATISFACTORY 


E 


NOTES: 
1. TO ESTABLISH COMBINED FAN CURVE, THE 
COMBINED TOTAL PR RE IS THE SUM 
OF INDIVIDUAL FAN PRESSURES AT EQUAL 
AIR FLOW RATES, LESS THE PRESSURE LOSS 
THE FAN CONNECTIONS. 


TOTAL PRESSUR 


AIR FLOW RATE THROUGH EACH FAN WILL BE 
THE SAME, SINCE AIR 1S CONSIDERED 
INCOMPRESSIBLE. 


~ FLOW RATE 
3. SYSTEM CURVE MUST INTERSECT 

TWO DIFFERENT FANS COMBINED FAN CURVE OX LARGE FLOW RATE 

UNSATISFACTORY FAN MAY HANDLE MORE AIR ALONE, 


WHEN SYSTEM CURVE DOES NOT INTERSECT 
COMBINED FAN CURVE, OR CROSSES PROJECTED 
COMBINED CURVE BEFORE FAN B CURVE, FAN 8B 
WILL MOVE MORE AIR THAN FAN A AND 8 IN 
SERIES. 


AMERICAN CONFERENCE | FANS 
OF GOVERNMENTAL SEES OPERATION 
INDUSTRIAL HYGIENISTS fates 96d “On (nee ee 


6-24 Industrial Ventilation 


js+— DESIGN PRESSURE | 


r NN 
* 


SYSTEM EFFECT LOSS 


) 


x 7 
3) 


FAN CATALOG 


N CURVE 
N 
_ 
\ ACTUAL PERFORMANCE 
OF FAN BECAUSE OF 
DEFICIENT ‘\ n i 
YST 
SeRPORNEE ‘ SYSTEM EFFECT 


FIGURE 6-14. System effect factor 


"R." To find the system effect correction value in inches water 
gauge, we use the fan inlet velocity with Figure 6-27. Since 
the duct area is 3.142 ft?, the velocity is 1592 fpm (5000 cfm 
+ 3.142 ft? = 1592 fpm). From Figure 6-27 we get a correction 
value of 0.19 “wg. This 0.19" value is added to the fan static 
pressure when selecting the fan from the multi-rating table. 
Select the fan for a static pressure of 8.19 "wg. Interpolating 
in Table 6-1, we find a selection for 5000 cfm and 8.19 "wg 
at 2002 rpm and 13.25 hp. This selection for a fan with an 
elbow at the inlet will result in operation at 5000 cfm and 8 
"we drawing 13.25 hp. 


i a ! 
\ Ne ; vy) f 
XS xy 


FIGURE 6-15. Fan discharge conditions 


ht DESIGN FLOW RATE i 


Note: The system effect coefficient compensates for the 
affect on the fan of an irregular air stream. This system 
effect coefficient is taken in addition to the friction loss 
used to calculate the system loss (Figure 5-13.) 


Figure 6-15 illustrates typical discharge conditions and the 
losses which may be anticipated. The magnitude of the change 
in system performance caused by elbows and other obstruc- 
tions placed too close to a fan inlet or outlet can be estimated 
for the conditions shown on Figures 6-18 through 6-25. 


FACTOR @ DESIGN FLOW RATE 


SN 
“-R 1220 mm 


ates e- 610 mm 


FIGURE 6~16. Inlet elbow 


Addition to system static pressure is given by reference to 
lettered curves in all but Figure 6-23. The additional static 
pressure, in "wg, is determined by obtaining the appropriate 
system effect coefficient from Figure 6-26 or 6-27 and mul- 
tiplying it by the fan inlet or discharge velocity pressure. 


A vortex or spin of the air stream entering the fan inlet may 
be created by non-uniform flow conditions as illustrated in 
Figure 6-24. These conditions may be caused by a poor inlet 
box, multiple elbows or entries near the inlet, or by other 
spin-producing conditions. Since the variations resulting in 
inlet spin are many, no System Effect Coefficients are tabu- 
lated. Where a vortex or inlet spin cannot be avoided or is 
discovered at an existing fan inlet, the use of turning vanes, 
splitter sheets, or egg crate straighteners will reduce the effect. 


6.4.2 Inspection and Maintenance: Material accumula- 
tion or abrasive wear on an impeller can cause a fan to "go 
out of balance." This unbalance will cause fan vibration. This 
may result in damage to or failure of the fan impeller, housing, 
bearings, or pedestal. Periodic cleaning and rebalancing of 
fans operating in air streams handling high material concen- 
trations is recommended. 


Regular observation of fan vibration levels can detect prob- 
lems before they develop to a damaging amplitude (see Figure 
6-17). Modern maintenance equipment permits the inspector 
to record vibration spectra. Review of changes in these spectra 
taken over time can indicate specific areas of developing 
problems with bearings, balance, belts or motors. Electronic 
or computerized vibration monitors are available to mount on 
fans used in critical operations. These devices can be set up 
with automatic alarm functions and/or to provide continuous 
information about a unit’s vibration level. 


It is not uncommon, during fan installation or motor/starter 
maintenance, for the fan impeller rotation direction to be 


Fans 6-25 


inadvertently reversed. Since fans do move a fraction of their 
rated capacity when running backward, incorrect rotation 
often goes unnoticed in spite of less effective performance of 
the exhaust system. 


Scheduled inspection of fans is recommended. Items 
checked should include: 


1. Bearings for proper operating temperature (lubricate 
them on the manufacturer’s recommended schedule). 


. Excessive vibration of bearings or housing. 
. Belt drives for proper tension and minimum wear. 
. Correct coupling or belt alignment. 


. Fan impeller for proper alignment and rotation. 


N vw BW NY 


. Impeller free from excess wear or material accumula- 
tion. 


7. Tight fan hold-down bolts. 
8. Tight fan impeller set screws or bushings. 
9. Proper installation of safety guards. 


Standard lockout/tagout procedures should be observed 
when servicing fan equipment or its associated duct. The 
electrical supply must be shut off and locked out at a discon- 
nect near the fan. When opening access doors or reaching into 
the fan inlet or outlet, the fan must be mechanically locked 
out by blocking the impeller from rotating. A warning tag 
should be used when blocking a fan. Do not open an access 
door while the fan is operating or coasting down. 


BE SURE to remove any inserted obstructions used to block 
impeller rotation when servicing is complete. 


REFERENCES 


6.1. Air Movement and Control Association, Inc.: AMCA 
Publication 201-90, Fans and Systems. AMCA, Ar- 
lington Heights, IL (1990). 


6.2. Gibson, N.; Lloyd, F.C.; Perry, G.R.: Fire Hazards in 
Chemical Plants from Friction Sparks Involving the 
Thermite Reaction. Symposium Series No. 25. Insn. 
Chem. Engrs., London (1968). 


6.3. Air Movement and Control Association, Inc.: AMCA 
Publication 99-86, Standards Handbook. AMCA, Ar- 
lington Heights, IL (1986). 

6.4. American Society of Heating, Refrigeration, and Air- 
Conditioning Engineers, Inc.: 1993 ASHRAE Hand- 
book, Fundamentals Volume. ASHRAE, Atlanta, GA 
(1993). 


6-26 Industrial Ventilation 


Vioration Frequency (CPM) 


oe fae ae 
Oc 
oO ¢ 


OQ 
oO 


~~ 100000 


Values shown are for 
fiitered readings taken jon 
the machine structure 
—| bearing cap 


628 in/sec 


in/sec 


in /sec 


in/Sec 


Vibration Velocity 


O.C1 | J+} tL} NL 
to) sa acres a oe clkareoaat aaa CN ee 
0.006 


0.004 pH ever bok os oe Ee | . DNL iNn/Sec 


0.003 | 


0.002 -—— Fe ac eee ee ee | LEN in/ sec 


0,004 


0.0049 in/sec 


Reprinted with the permission of IRB Mechanalysis, Inc. 


A ME R i CA N ¢ O NF ERE NC E | MACH. VIBRATION 


VERITY CHART 
OF GOVERNMENTAL ee en 


INDUSTRIAL HYGIENISTS iis 7006] 205 eu Ge 


Fans 6-27 


BLAST AREA 


CUTOFF OUTLET AREA 
CENTRIFUGAL 


FAN a 


DISCHARGE DUCT 


L 100% EFFECTIVE DUCT LENGTH 


TO CALCULATE 100% EFFECTIVE DUCT LENGTH, ASSUME A MINIMUM OF 2-1/2 DUCT DIAMETERS FOR 2500 
FPM OR LESS. ADD 1 DUCT DIAMETER FOR EACH ADDITIONAL 1000 FPM. 


EXAMPLE: 5000 FPM = 5 EQUIVALENT DUCT DIAMETERS. IF THE DUCT IS RECTANGULAR WITH SIDE 
DIMENSIONS a AND b, THE EQUIVALENT DUCT DIAMETER IS EQUAL TO (4ab/x)®5 


12% 
Effective 
Duct Duct 


25% 
Effective 
Duct 


50% 
Effective 
Ouct 


100% 
Effective 
Ouct 


ee eee 
System Effect Curve 


Blast Area_ 
Outlet Area 


DETERMINE SEF BY USING FIGURE 6-26 OR 6-27 


Reprinted from AMCA Publication 201-90, FANS AND SYSTEMS, by 
permission of the Air Movement and Control Association, Inc.©") 


AMERICAN CONFERENCE FOR OUTLET DUCTS— 


OF GOVERNMENTAL — | CENTRIFUGAL FANS | 


| IN 


6-28 Industrial Ventilation 


a AXIAL FAN 


L 100% EFFECTIVE DUCT LENGTH : 


TO CALCULATE 100% EFFECTIVE DUCT LENGTH, ASSUME A MINIMUM OF 2-1/2 DUCT 
DIAMETERS FOR 2500 FPM OR LESS. ADD 1 DUCT DIAMETER FOR EACH ADDITIONAL 
1000 FPM. 


EXAMPLE: 5000 FPM = 5 EQUIVALENT DUCT DIAMETERS 


50% 
Effective 


100% 
Effective 
Duct 


DETERMINE SEF BY USING FIGURE 6-26 OR 6-27 


Reprinted from AMCA Publication 201-90, FANS AND SYSTEMS, by 
permission of the Air Movement and Control Association, Inc.©") 


SYSTEM EFFECT CURVES | 
FOR OUTLET DUCTS- 
AXIAL FANS 


AMERICAN CONFERENCE 
/ OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


Fans 6-29 


POSITION C 


POSITION D 


a 
a ~ 


N 
7 
vl POSITION B 
oy. 4 
Zz 
a 
a we 


POSITION A 


DETERMINE SEF BY USING FIGURES 6-26 AND 6-18 


For DWDI fans determine SEF using the curve for 
SWSI fans. Then apply the appropriate multiplier 
from the tabulation below 


MULTIPLIERS FOR DWDI FANS 


A N P.Q 
B M-N oP 
L-M N 
L-M N 
R 
Q 


ELBOW POSITION A = AP X 1.00 
ELBOW POSITION B = AP X 1.25 
ELBOW POSITION C = AP X 1.00 
ELBOW POSITION D = AP X 0.85 


7 
< 


man 
nana 


oO 

ee 

005% | 0008 | zz25 

za oome | 
<|oore 


NO SYSTEM EFFECT FACTOR 


na 
4 


HEE 


QOAT 
44 


Reprinted from AMCA Publication 201-90, FANS AND SYSTEMS, by 
permission of the Air Movement and Control Association, Inc.) 


FOR OUTLET ELBOWS 
ON CENTRIFUGAL FANS 


a ae a a 


AMERICAN CONFERENCE 
| OF GOVERNMENTAL | 
| INDUSTRIAL HYGIENISTS 


DATE 


6-30 industrial Ventilation 


LENGTH — 
OF DUCT | | (0 SYSTEM EFFECT FACTORS 


NO 20 8D 
DUCT DUCT DUCT 


P R-S 


A. TWO-PIECE MITERED 90° ROUND SECTION ELBOW NOT VANED 


LEN NGTH fg gt jie ee giticnsh Me ae 
OF DUCT | . SYSTEM EFFECT FACTORS 


NO 2D 5D 
DUCT DUCT DUCT 

@) Q 

Q R=-S 

R ners 

R-S 

S 


B. THREE PIECE METERED 90° ROUND SECTION ELBOW -- NOT VANED 


ENG Te 4 - SYSTEM EFFECT FACTORS 


NO 2D 3D 
DUCI DUCT BUCT 


; 
U 

U-V 
U-V 
V-W 


C. FOUR OR MCRE PIECE MITERED 90° ROUND SECTION ELBOW ~-— NOT VANED 


D=Diameter of the inlet collar. 
The inside area of the square duct (H X H) should be equal to the inside area of the fan inlet collar. 
+The maximum permissible angle of any converging elemen: of the transistion is 15°, and for a diverging element 7. 


Reprinted from AMCA Publication 201-76. FANS AND SYSTEMS, by 

ermission of the Air movement ond Control Association, Inc. (6.1) 

eet ee ee | SYSTEM EFFECT CURVES FOR 
AMERICAN CONFERENCE VARIOUS MITERED ELBOWS 
COVERNMENTAL WITHOUT TURNING VANES 


[ INDus STRIAL HYGIENISTS AE FT 


Fans 6-31 


LENGTH SYSTEM EFFECT FACTORS 


ee | NO 20 50 
DUCT DUCT DUCT 


SYSTEM EFFECT FACTORS 
NO 2D SD 
DUCT DUCT DUCT 


B. SQUARE ELBOW WITH INLET TRANSITION ~- 3 LONG TURNING VANES. 


LENGTH __ ca | 
OF DUCT 1 | | SYSTEM EFFECT FACTORS 


NC 2 5D 
DUCT 


C. SQUARE ELEOW WHTH INLET TRANSIHFCN -—~- SHORT TURNING VANES. 


THE INSIDE AREA OF THE SQUARE DUCT (H X H) iS EQUAL TO THE INSIDE AREA CIRCUMSCRIBED BY THE FAN INLET COLLAR, 
THE MAXIMUM PERMISSIBLE ANGLE OF ANY CONVERGING ELEMENT OF THE TRANSITION IS 15°, AND FOR A DIVERGING 
ELEMENT 7.5, 


Reprinted from AMCA Publication 201-76. PANS AND SYSTEMS, by 
permission of the Air movement and Control Association, Inc. (6.1) 


AMERICAN CONFERENCE SYSTEM BEFECT CURVES 


a. CAA ON R GT POR OUTLET DOUCTS= 
O k GO V EI RNMEN [ Als L FANS 


| INDUSTRIAL | ISTS Samat Se 


6-32 Industrial Ventilation 


LENGTH 
OF DUCT — 


SYSTEM EFFECT FACTORS# 


NO 20 
DUCT 


A. NON-UNIFORM FLOW INTO A FAN INLET 

BY A 90° ROUND SECTION ELBOW ~ NO THE REDUCTION IN FLOW RATE AND PRESSURE FOR 

TURNING VANES. THIS TYPE OF INLET CONDITION IS IMPOSSIBLE TO 
TABULATE. THE MANY POSSIBLE VARIATIONS IN 
WIDTH AND DEPTH OF THE DUCT INFLUENCE THE 
REDUCTION IN PERFORMANCE TO VARYING DE~ 
GREES AND THEREFORE THIS INLET SHOULD BE 
AVOIDED. FLOW RATE LOSSES AS HIGH AS 45% 
HAVE BEEN OBSERVED. EXISTING INSTALLATIONS 
CAN BE IMPROVED WITH GUIDE VANES OR THE 
CONVERSION TO SQUARE OR MITERED ELBOWS 
WITH GUIDE VANES. 


*Values shown are in modification of the 
original chart, 


B. NON-UNIFORM FLOW INDUCED INTO FAN 


Reprinted from AMCA Publication 201-76. FANS AND SYSTEMS, by 
ermission of the Air movement and Control Association, Inc. 


AMERICAN CONFERENCE | NON- UNIFORM INLET 
OF GOVERNMENTAL FLOWS 
INDUSTRIAL. HYGIENISTS 


Fans 6-33 


Ls and J Hl Je GE 
TY ayy 
a vee 


I turning | | yer OS 


VANES oa | aad LL | 
a SS 7 
[eX yl. setae we S | . WHat 


| A ; ROTATION SON y ~ 7 “ROTATION 
a e SN ane 
i Jt =_— Dae SSH 


TURNING 
VANES 

CORRECTED PRE~ 

ROTATING SWIRL 


—— Foyt ea | poo sans 
K li | | 


CORRECTED COUNTER- 
ROTATING SWIRL 


C. NON-UNIFORM FLOW INTO A FAN INLET BY AN INDUCED VORTEX, SPIN OR SWIRL. 


Reprinted from AMCA Publication 201-90 FANS AND SYSTEMS by 
permission of the Air Movement and Control Association Inc. (6.1) 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


NON— UNIFORM INLET 
CORRECTIONS 


6-34 Industrial Ventilation 


3 


A. FREE INLET AREA PLANE —— FAN WITH INLET COLLAR. 
POINT OF TANGENT 
WITH FAN HOUSING SIDE 
AND INLET CONE RADIUS 


INLET PLANE 


INLET PLANE 


B. FREE INLET AREA PLANE —~ FAN WITHOUT INLET COLLAR. 


PERCENTAGE OF UNOBSTRUCTED 
INLET AREA 


SYSTEM EFFECT FACTORS 


NO LOSS 


DETERMINE SEF BY CALCULATING INLET VELOCITY AND USING FIGURE 6-26 


Reprinted from AMCA Publication 210-90, FANS AND SYSTEMS, by 
permission of the Air and Contro! Association Inc. (6.1) 


FOR INLET OBSTRUCTIONS 


— S55 [roo a a sa : | 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
f INDUSTRIAL HYGIENISTS 


Fans 6-35 


Loss Factor Equivalents 
for System Effect Curvese 


Curve | Fsyg { Curve 


F 16.0 P 

G 14.3 Q 1.60 
H 12.8 R 1.20 
| Tle S 0.80 
J 9.62 T 0.53 


To use this table: 1) Obtain the curve letter from Figures 
6—-18 through 6-22 or Figure 6-25. 
2) For inlet system effects, multiply the 
equivalent loss coefficient from the above 

table by the fan inlet velocity pressure. 

3) For outlet system effects, multiply the 
equivalent loss coefficient from the above 
table by the fan outlet velocity pressure. 


*Fsys Values are in number of velocity pressures. 
refer to Figure 6-27. 


For loss directly in "Wa, 


AMERICAN CONFERENCE 
» OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


SYSTEM EFFECT CURVES | 


6-36 Industrial Ventilation 


FGHI J KL M 


{\ 


O 
= 
ea) 
a 
Ss 
2 
ui 
(ace 
— 
Y 
ea) 
Lid 
ow 
Oo 
| 
a 
Oo 
—_ 
oO 
< 
Le 
— 
O 
bd 
tL 
in 
ud 
= 
Lid 
fb 
Y 
> 
a) 


SD 6G. de 82910 1S 20 eS 30 40 SO 60 


AIR VELOCITY, FPM IN HUNDREDS 
(Air Density = 0.075 lIbs/ft>) 


f*Enter the chart at the appropriate air velocity (on the abcissa) read up to the applicable ff 
curve, then across from the curve (to the ordinate) to find the SEF at standar : 
Fair density. 

«Adapted for metric from AMCA Publication 201-90, FANS AND SYSTEMS, by permission 

of the Air Movement and Control Association, Inc. (6. 


| AMERICAN CONFERENCE | 
| OF GOVERNMENTAL 
: INDUSTRIAL HYGIENISTS — BGS Sennen 1ST ae GraF = 


SYSTEM EFFECT CURVES 


Chapter 7 
REPLACEMENT AND RECIRCULATED AIR 


7.1. INTRODUCTION ............-...-.. 
7.2 REPLACEMENTAIR................. 
7.3. REPLACEMENT AIR DISTRIBUTION ...... .- 
7.4 REPLACEMENT AIR FLOW RATE. ......... 7-5 


7.5 ROOM PRESSURE 


7.6 ENVIRONMENTAL CONTROL ........... 
7.7 ENVIRONMENTAL CONTROL AIR FLOW RATE . 7-6 


7AL 
712 
7.13 


REPLACEMENT AIR HEATING EQUIPMENT. . . 
COST OF HEATING REPLACEMENT AJR... .. 
AIR CONSERVATION 


7.13.1 Reduced Flow Rate 
7.13.2 Untempered Air Supply 
7.13.3 Energy Recovery 
7.13.4 Selection of Monitors 


7.8 AIRCHANGES ...............0004.- 7-6 7.14 EVALUATION OF EMPLOYEE EXPOSURE 
7.9 AIRSUPPLY TEMPERATURES .......... 7-6 LEVELS .. 1... ee eee 
7.10 AIR SUPPLY VS. PLANT HEATING COSTS . 7-8 REFERENCES... 2.2) ee 
Figure 7-1 Cold Zones Vs. Overheated Zones ........ 7-2 Figure 7-9 By-Pass Steam System .........-0.. 
Figure 7-2 How Fan Performance Falls Off Figure 7-10 — Integral Face and By-Pass Coil... 2... 
Under Negative Pressure... 2. ......40. 7-3 Figure 7-11 Indirect Fired Unit... 0... ee, 
Figure 7-3 Relationship Between Air Pressure and Figure 7-12 Direct Fired Unit .. 2... ........0.. 
Amount of Force... 2... 0. ee 7-4 Figure 7-13 Direct Fired By-Pass Unit... .....00.. 
Figure 7-4 Throw Patterns and Distance From Different Figure 7-14 — Recirculation Decision Logic... 2.2.2... 
Register Adjustments... ............ 7-7 Figure 7-15 Schematic Diagram of Recirculation 
Figure 7-5 Seasonal Air Ventilation ............. 7-8 Monitoring System... 2. ...0..0.-240. 
Figure 7-6 Single Coil Steam Unit... ..........0. 7-9 Figure 7-16 Schematic of Recirculation From Air Cleaning 
Figure 7-7 Steam Coil Piping. 2. 2. ee ee 7-10 Devices (Particulates) ............. 
Figure 7-8 Multiple Coil Steam Unit. 2... 0. 22 2... 7-11 


7-2 industrial Ventilation 


7.1 INTRODUCTION 


Chapters | through 6 describe the purpose, function, and 
design of industrial exhaust systems. As mentioned in Chapter 
1, Section 1.2, supply systems are used for two basic purposes: 
to create a comfortable environment and to replace air ex- 
hausted from the building. It is important to note that while 
properly designed exhaust systems will remove toxic con- 
taminants, they should not be relied upon to draw outdoor air 
into the building. If the amount of replacement air supplied 
to the building is lower than the amount of air exhausted, the 
pressure in the building will be lower than atmospheric. This 
condition is called "negative pressure" and results in air 
entering the building in an uncontrolled manner through 
window sashes, doorways, and walls. In turn, this may lead 
to many undesirable results such as high velocity drafts, 
backdrafting, difficulty in opening doors, etc. 


To minimize these effects, design the mechanical supply 
systems to introduce sufficient outside air to avoid excessive 
negative or positive pressure conditions. A properly designed 
and installed air supply system can provide both replacement 
air and effective environmental control. Provided that impor- 
tant health and safety measures are taken, recirculation of the 
exhaust air may be an effective method that can substantially 
reduce heating and/or cooling costs. 


7.2 REPLACEMENT AIR 


Air will enter a building in an amount to equal the flow rate 
of exhaust air whether or not provision is made for this 
replacement. However, the actual exhaust flow rate will be 
less than the design value if the plant is under negative 
pressure. If the building perimeter is tightly sealed, thus 
blocking effective infiltration of outdoor air, a severe decrease 
of the exhaust flow rate will result. If, on the other hand, the 
building is relatively old with large sash areas, air infiltration 
may be quite pronounced and the exhaust system performance 
will decrease only slightly and other problems may occur. 


| EXHAUST : 


S 3 
< ee | = 
= im = 
J ea) wn] 

A ; ar) 


5 OVER HEATED COLD 
ZONE ZONE ZONE 


Figure 7-1. Under negative pressure conditions, workers in the cold zones 
turned up thermostats in an attempt to get heat. Because this did nothing 
to stop leakage of cold air, they remained cold while the center of plant was 
overheated. 


TABLE 7-1. Negative Pressures and Corresponding Velocities 
Through Crack Openings (Calculated with air at room temperature, 
standard atmospheric pressure, C, = 0.6) 


Negative Pressure, "wg Velocity, fpm 
0.004 150 
0.008 215 
0.010 240 
0.014 285 
0.016 300 
0.018 320 
0.020 340 
0.025 380 
0.030 415 
0.040 480 
0.050 540 
0.060 590 
0.080 680 
0.100 760 
0.150 930 
0.200 1080 
0.250 1200 
0.300 1310 
0.400 1520 
0.500 1700 
0.600 1860 


When the building is relatively open, the resultant in-plant 
environmental condition is often undesirable since the influx 
of cold outdoor air in the northern climates chills the perimeter 
of the building. Exposed workers are subjected to drafts, space 
temperatures are not uniform, and the building heating system 
is usually overtaxed (see Figure 7-1). Although the air may 
eventually be tempered to acceptable conditions by mixing as 
it moves to the building interior, this is an ineffective way of 
transferring heat to the air and usually results in fuel waste. 


Experience has shown that replacement air is necessary for 
the following reasons: 


1. To insure that exhaust hoods operate properly. A \ack 
of replacement air and the attendant negative pressure 
condition results in an increase in the static pressure 
the exhaust fans must overcome. This can cause a 
reduction in exhaust flow rate from all fans and is 
particularly serious with low-pressure fans such as 
wall fans and roof exhausters (see Figure 7-2). 


2. To eliminate high-velocity cross-drafts through win- 
dows and doors. Depending on the negative pressure 
created, cross drafts may be substantial (see Table 7-1). 
Cross-drafts not only interfere with the proper opera- 


Replacement and Recirculated Air 7-3 


NEGATIVE PRESSURE IN BUILDING 


— ORIGINAL SYSTEM 
y~ PROPELLER FAN 


Va 


ROOM NEGATIVE PRESSURE 


ci mt—— LARGE FLOW LOSS 


NEGATIVE PRESSURE IN BUILDING \ 


Dent 


SYSTEM 


_ CENTRIFUGAL. 
FAN 


ROOM NEGATIVE PRESSURE 


Ee SMALL FLOW LOSS 


< 


HOW FAN PERFORMANCE FALLS | 
OFF UNDER NEGATIVE PRESSURE | 


FIGURE ‘_.¢ 


| AMERICAN CONFERENCE 
|} OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


Tod 


Industrial Ventilation 


TABLE 7-2. Negative Pressures Which May Cause Unsatisfactory Conditions Within Buildings 


Negative Pressure "wg 


Adverse Conditions 


0.01 to 0.02 
0.01 to 0.05 


Worker Draft Complaints — High velocity drafts through doors and windows. 
Natural Draft Stacks Ineffective — Ventilation through roof exhaust ventilators, flow through 


stacks with natural draft greatly reduced. 


0.02 to 0.05 


Carbon Monoxide Hazard — Back drafting will take place in hot water heaters, unit heaters, 


furnaces, and other combustion equipment not provided with induced draft. 


0.03 to 0.10 


General Mechanical Ventilation Reduced — Air flows reduced in propeller fans and low 


pressure supply and exhaust systems. 


0.05 to 0.10 
0.10 to 0.25 


tion of exhaust hoods, but also may disperse contami- 
nated air from one section of the building to another 
and can interfere with the proper operation of process 
equipment such as open top solvent degreasers. In the 
case of dusty operations, settled material may be dis- 
lodged from surfaces and result in recontamination of 
the workroom. 


. To insure operation of natural draft stacks such as 


combustion flues. Moderate negative pressures can 
result in backdrafting of flues which may cause a 
dangerous health hazard from the release of combus- 
tion products, principally carbon monoxide, into the 
workroom. Back drafting may occur in natural draft 
stacks at negative pressures as low as 0.02 "wg (see 
Table 7-2). Secondary problems include difficulty in 
maintaining pilot lights in burners, poor operation of 
temperature controls, corrosion damage in stacks and 
heat exchangers due to condensation of water vapor in 
the flue gases. 


. To eliminate cold drafts on workers. Drafts not only 


cause discomfort and reduce working efficiency but 
also may result in lower overall ambient temperatures. 


. To eliminate differential pressure on doors. High dif- 


ferential pressures make doors difficult to open or shut 
and, in some instances, can cause personnel safety 
hazards when the doors move in an uncontrolled fash- 
ion (see Figure 7-3 and Table 7-2). 


. To conserve fuel. Without adequate replacement air, 


uncomfortable cold conditions near the building pe- 
rimeter frequently lead to the installation of more 
heating equipment in those areas in an attempt to 
correct the problem. These heaters take an excessive 
amount of time to warm the air and the over-heated air 
moving toward the building interior makes those areas 
uncomfortably warm (see Figure 7-1). This in turn 
leads to the installation of more exhaust fans to remove 
the excess heat, further aggravating the problem. Heat 
is wasted without curing the problem. The fuel con- 
sumption with a replacement air heating system usu- 


Doors Difficult to Open — Serious injury may result from non-checked, slamming doors. 
Local Exhaust Ventilation Impaired — Centrifugal fan exhaust flow reduced. 


ally is lower than when attempts are made to achieve 
comfort without replacement air (see Section 7.10). 


7.3 REPLACEMENT AIR DISTRIBUTION 


Replacement air distribution is as critical as air volume 
(quantity) in industrial ventilation system design. Poor air distri- 
bution can destroy the control provided by a well-designed 
exhaust system. Non-turbulent air flow is particularly critical in 
indoor firing ranges, pharmaceutical plants, electronic compo- 
nents plants, some paint shops, and similar facilities. 


Designers often use the general heating, ventilating, and air 
conditioning system or plant air system to provide the supply 
(or replacement) air to replace air exhausted by local exhaust 
hoods. Unfortunately, plant air systems use high throw dif- 
fusers to mix the warmer/cooler air with the air already in the 
plant. Many times the throw distance is over 40 feet. This 
mixing effect causes turbulence near local exhaust systems. 
The local exhaust hoods must then be redesigned to draw in 
more air to control the contaminants and overcome the turbu- 
lence. This increases energy costs due to the need for larger 
fans and motors. Sometimes hoods cannot capture contami- 
nants and overcome turbulence. Hence, workers could still be 
overexposed even with a local exhaust system in place. There- 


(00) 22 104 -T 
TOME IEDs ope e 78 - Se eet 
O oO 
Oo 2, 1 
AO ae ER es 62a me! 
Zz door 
coe a ee, Ogee 20 sa ft 
=) <t 
WY 2 a 
Boo 8 Bo Jo i 
a 
ae 5. i Y 
. 25 + 1.3 a. te 26 +e 
50 +26 A 59 + 
Zz, °8 (T] 
ao 2 
Prd ar oO ~) a 
FF LO ee Oo 780-4 
1.00 + 5.2 ae xls 


Figure 7-3. Relationship between air pressure and amount of force needed 
to open or close an average-sized door 


fore, locate local exhaust hoods away from the turbulent 
effects of the plant air distribution system. 


If the supply air system does not sufficiently cool the 
employees, pedestal fans are often used. Pedestal fans also 
destroy contaminant control by causing turbulence near the 
local exhaust hood and should not be allowed. 


One method of providing non-turbulent air to the facility 
is to pass air through a supply air plenum built as part of the 
ceiling and/or through perforated duct. Cover the plenum face 
with perforated sheet metal. The ceiling plenum or duct runs 
should cover as large an area as possible to diffuse the air flow. 
A plenum wall providing cross-flow ventilation can be used 
only if the workers position themselves between the supply 
air system and the contaminant source and should not be used 
if the design velocity is over 100 fpm. 


Perforated drop-type ceilings work best in facilities with 
ceiling heights of less than 15 feet. Hoist tracks, lighting, and 
fire protection systems can be built into the ceiling. In some 
cases, fire protection will be required above and below the 
ceiling. Use perforated duct for ceilings over 15 feet. Perfo- 
rated duct manufacturers have computer programs to assist 
designers in determining duct sizes, shapes, and types as well 
as the location of pressure adjusting devices such as orifice 
plates and reducers. Air flow delivery in large bays may 
require supplemental air delivered at work stations to provide 
comfortable conditions for the workers. 


Feeding the air into the plenum is also critical. High veloc- 
ity flow into the plenum will cause the same turbulence prob- 
lems as the large diffusers commonly found in plants that 
attempt to throw the air to the floor. Consider feeding the 
plenum with perforated duct to diffuse the air inside the 
plenum. Another method of distributing air flow, from either 
a ceiling or wall-mounted plenum, is to design the plenum with 
two perforated plates, one fixed and one adjustable, located 
2-6 inches apart. Air flowing through slightly offset holes will 
encounter more resistance; thus air quantities passing through 
the low-flow areas will increase. The adjustable plates must be 
small enough to fine tune the air flow from the plenum. 


7.4 REPLACEMENT AIR FLOW RATE 


In most cases, replacement air flow rate should approxi- 
mate the total air flow rate of air removed from the building 
by exhaust ventilation systems, process systems and combus- 
tion processes. Determination of the actual flow rate of air 
removed usually requires an inventory of air exhaust locations 
and any necessary testing. When conducting the exhaust 
inventory, it is necessary not only to determine the quantity 
of air removed, but also the need for a particular piece of 
equipment. At the same time, reasonable projections should 
be made of the total plant exhaust requirements for the next 
one to two years, particularly if process changes or plant 
expansions are contemplated. In such cases it can be practical 
to purchase a replacement air unit slightly larger than immedi- 


Replacement and Recirculated Air 7-5 


ately necessary with the knowledge that the increased capac- 
ity will be required within a short time. The additional cost of 
a larger unit is relatively small and in most cases the fan drive 
can be regulated to supply only the desired quantity of air. 


Having established the minimum air supply quantity nec- 
essary for replacement air purposes, many plants have found 
that it is wise to provide additional supply air flow rate to 
overcome natural ventilation leakage and further minimize 
drafts at the perimeter of the building. 


7.5 ROOM PRESSURE 


While negative pressure can cause adverse conditions as 
described in Sections 7.1-7.3, there are situations where 
negative pressures are desired. An example is a room or area 
where a contaminant must be prevented from escaping into 
the surrounding area. It also may be desirable to maintain a 
room or area under positive pressure to maintain a clean 
environment. Either of these conditions can be achieved by 
setting and maintaining the proper exhaust/supply flow dif- 
ferential. Negative pressure can be achieved by setting the 
exhaust volumetric flow rate (Q) from the area to a level 
higher than the supply rate. A good performance standard for 
industrial processes is to set a negative pressure differential 
of 0.04 +/~ 0.02 "wg. Conversely, positive pressure is 
achieved by setting the supply air flow rate higher than the 
exhaust rate. The proper flow differential will depend on the 
physical conditions of the area, but a general guide is to set a 
5% flow difference but no less than 50 cfm. If the volume 
flows vary during either negatively or positively pressurized 
processes, it is easier to maintain the desired room pressure 
by adjusting the supply air. 


Some designers use transfer grilles and a pressure sensor 
in the room to maintain a desired room pressure. Air is allowed 
to seep from adjacent hallways, offices and other non-indus- 
trial areas. Do not use transfer grilles between areas where 
contaminant migration is possible. 


7.6 ENVIRONMENTAL CONTROL 


There are generally three types of industrial ventilation 
systems in most plants: 1) return air for the clean plant air; 2) 
areturn air system where low level contaminants are diluted 
with fresh air (dilution ventilation); and 3) contaminant-laden 
air drawn through a local exhaust hood or ventilation system. 
In addition to toxic contaminants which are most effectively 
controlled by hoods, industrial processes may create an unde- 
sirable heat load in the work space. Modern automated ma- 
chining, conveying, and transferring equipment require 
considerable horsepower. Precision manufacturing and as- 
sembling demand increasingly higher light levels in the plant 
with correspondingly greater heat release. The resulting in- 
plant heat burden raises indoor temperatures, often beyond 
the limits of efficient and healthful working conditions and, 
in some cases, beyond the tolerance limits for the product. 


7-6 Industrial Ventilation 


Environmental control of these factors can be accommo- 
dated through the careful use of the supply system. Industrial 
air conditioning may be required to maintain process specifi- 
cations and employee health. Many times the designer can use 
a setpoint higher than the 50-55 F used in conventional HVAC 
designs. ASHRAE gives basic criteria for industrial air con- 
ditioning in HVAC applications.” (It must be noted that 
radiant heat cannot be controlled by ventilation and methods 
such as shielding, described in Chapter 2, are required.) 
Sensible and latent heat released by people and the process 
can be controlled to desired limits by proper use of ventilation. 


The HVAC industry uses automated building contro! and 
direct digital control (DDC) in many facilities. The technol- 
ogy can be applied to industrial ventilation with careful plan- 
ning. DDC uses computers and microprocessors tied to 
sensors and actuators to form a feedback and control system. 
DDC can be useful in industrial ventilation systems to control 
temperature, humidity, and relative room pressures. DDC 
systems can also track the system performance at hoods, fans, 
heating and cooling, and air pollution control equipment. 
DDC is especially useful in preventive maintenance. How- 
ever, DDC systems for industrial ventilation systems are 
complicated. Many are "one-of-a-kind" systems designed by 
a controls manufacturer and they require trained personnel to 
operate. 


Many industrial processes release minor amounts of "nui- 
sance" contaminants which, at low concentrations, have no 
known health effects but which are unpleasant or disagreeable 
to the workers or harmful to the product. The desire to provide 
a clean working environment for both the people and the 
product often dictates controlled air flow between rooms or 
entire departments. Evaluate the air streams returned into the 
facility to determine if the air pollution control devices (e.g., 
filters, cyclones) provide sufficient cleaning to prevent em- 
ployee exposure to "nuisance" contaminants. In addition, 
systems with known contaminants require controls listed in 
Section 7.12 and 7.13. The facility must employ trained 
mechanics and support a preventive maintenance program to 
sufficiently protect the workers. 


7.7 ENVIRONMENTAL CONTROL AIR FLOW RATE 


The design supply air flow rate depends on several factors 
including the health and comfort requirements. Sensible heat 
can be removed through simple air dilution (see Chapter 2 
under ventilation). 


"Nuisance" or undesirable contaminants can also be re- 
duced by dilution with outdoor air. The control of odors from 
people at various conditions of rest and work can be accom- 
plished with the outdoor air flow rate described in Chapter 2. 
However, these data apply mainly to offices, schools and 
similar types of environment and do not correspond well with 
the usual industrial or commercia! establishment. Experience 
shows that when the air supply is properly distributed to the 


TABLE 7-3. Air Exchanges Vs. Room Sizes 


Air changes/ Air changes/ 
Room Size Room ft? minute hour 
40x40x12high 19,200 11,650/19,200 = 0.61 36 
40 x 40 x 20 high 32,000 11,650/32,000 = 0.364 22 


working level (i.e., in the lower 8-10 ft of the space), outdoor 
air supply of 1—2 cfm/ft? of floor space will give good results. 
Specific quantities of outdoor air must be obtained from 
criteria developed by groups such as ASHRAE. 


7.8 AIR CHANGES 


"Number of air changes per minute or per hour" is the ratio 
of the ventilation rate (per minute or per hour) to the room 
volume. "Air changes per hour" or "air changes per minute" 
is a poor basis for ventilation criteria where environmental 
control of hazards, heat, and/or odors is required. The required 
ventilation depends on the problem, not on the size of the 
room in which it occurs. For example, let us assume a situation 
where 11,650 cfm would be required to control solvent vapors 
by dilution. The operation may be conducted in either of two 
rooms, but in either case, 11,650 cfm is the required ventila- 
tion. The "air changes," however, would be quite different for 
the two rooms. As can be seen in Table 7-3, for the same "air 
change" rate, a high ceiling space will require more ventilation 
than a low ceiling space of the same floor area. Thus, there is 
little relationship between “air changes" and the required 
contaminant control. 


The "air change" basis for ventilation does have some 
applicability for relatively standard situations such as office 
buildings and school rooms where a standard ventilation rate 
is reasonable. It is easily understood and reduces the engineer- 
ing effort required to establish a design criteria for ventilation. 
It is this ease of application, in fact, which often leads to lack 
of investigation of the real engineering parameters involved 
and correspondingly poor results. 


7.9 AIR SUPPLY TEMPERATURES 


Supply air temperature is controlled by the demand for 
heating and cooling. Factors to consider in maintaining a 
comfortable work environment for occupants are: setpoint 
temperature, humidity control, air distribution, and air flow 
rate. Where high internal heat loads are to be controlled, 
however, the temperature of the air supply can be appreciably 
below that of the space by reducing the amount of heat 
supplied to the air during the winter months and by deliber- 
ately cooling the air in the summer. When a large air flow rate 
is delivered at approximately space temperatures or somewhat 
below, the distribution of the air becomes vitally important in 
order to maintain satisfactory environmental conditions for 
the persons in the space. 


Maximum utilization of the supply air is achieved when the 


air is distributed in the "living zone" of the space, below the 
8-10 foot level (see Figure 7-4). When delivered in this 
manner — where the majority of the people and processes are 
located — maximum ventilation results with minimum air 
handling. During the warm months of the year, large air flow 
in the working space at relatively high velocities is welcomed 
by the workers. During the winter months, however, care must 
be taken to insure that air velocities over the person, except 
when extremely high heat loads are involved, are kept within 
acceptable values (see Chapter 2, Table 2-5). To accomplish 
this, the air can be distributed uniformly in the space or where 
required for worker comfort. Heavy-duty, adjustable, direc- 
tional grilles and louvers have proven to be very successful in 
allowing individual workers to direct the air as needed.7) 
Light gauge, stamped grilles intended for commercial use are 
not satisfactory. Suitable control must be provided to accom- 
modate seasonal and even daily requirements with a minimum 
of supervision or maintenance attention. 


Chapter 2 describes the relative comfort that can be derived 
through adequate air flow control. Published tables of data by 
register and diffuser manufacturers indicate the amount of throw 
(projection) and spread that can be achieved with different 
designs at different flow rates (see Figure 7-4). Terminal veloci- 
ties at the throw distance can also be determined. 


va) 
yo” Wet 
A’ DEFLECTIO 
O 4 : nwa 
a 10" 2030 


Ry 


"E” DEFLECTION 


PLAN VIEW 


Pk 


UP PROJECTION 


SIDE VIEW 


POTATTTT 
PLEEL EEL eae 
AIGHT 

Q S 

AN 

a = 
ms 
\ 
al 
SN 


Replacement and Recirculated Air 7-7 


Multiple point distribution is usually best since it provides 
uniformity of air delivery and minimizes the re-entrainment 
of contaminated air that occurs when large volumes are 
"dumped" at relatively high velocities. Depending on the size 
and shape of the space and the amount of air to be delivered, 
various distributional layouts are employed. Single point 
distribution can be used; however, it is usually necessary to 
redirect the large volume of air with a baffle or series of baffles 
in order to reduce the velocity close to the outlet and minimize 
re-entrainment. In determining the number and types of reg- 
isters or outlet points, it also is necessary to consider the effect 
of terminal air supply velocity on the performance of local 
exhaust hoods. 


When large amounts of sensible heat are to be removed 
from the space during the winter months, it is most practical 
to plan for rapid mixing of the cooler air supply with the 
warmer air in the space. During the summer months, the best 
distribution usually involves minimum mixing so that the air 
supply will reach the worker at higher velocities and with a 
minimum of heat pickup. These results can be obtained by 
providing horizontal distribution of winter air over the 
worker’s head, mixing before it reaches the work area and 
directing the air toward the worker through register adjust- 
ment for the summer months (see Figure 7-5). 


° 
{ow 
N 


| o 
a eo 
TN] 


°42 


55°42° 22°0°22 
“| 


G DEFLECTION 


: = ep eA 


HORIZONTAL PROJECTION DOWN PROJECTION 


FIGURE 7-4. Throw patterns and distance from different register adjustments (REF. 7-2) 


7-8 Industrial Ventilation 


Delivered air temperatures during the winter usually range 
from 65 F—-68 F for work areas without much process heat or 
vigorous work requirement downward to 60 F or even 55 F 
where hard work or significant heat sources are involved. For 
summer operation, the temperature rise in indoor air can be 
estimated as described in Chapter 2. Evaporative cooling 
should be considered for summer operation. Although not as 
effective as mechanical refrigeration under all conditions, 
evaporative cooling significantly lowers the temperature of 


“S\N PULL CHAIN FOR 


10’ APPROX. FAST ADJUSTMENT 


WINTER ~ LOW AIR MOTION 
IN WORKING ZONE 


the outdoor air even in humid climates, improves the ability 
of the ventilation air to reduce heat stress, and costs much less 
to install and operate. 


7.10 AIR SUPPLY VS. PLANT HEATING COSTS 


Even if the supply air were drawn into the building simply 
by the action of the exhaust fans, during the winter months 
there will be an added burden on the plant heating system and 
fuel costs will rise. Experience has shown, however, that when 


SIDE WALL GRILLE 


SUMMER — HIGH AIR MCTION 


IN WORKING ZONE 


See PULL ROD FOR 


FAST ADJUSTMENT 


CEILING 


WINTER ~ LOW AIR MOTION 
IN WORKING ZONE 


OUT GE YT 


SUMMER — HIGH AIR MOTION 
IN WORKING ZONE 


FIGURE 7-5. Seasonal air ventilation 


~~ STEAM COIL 


lia 
Law 
oO) 7 


‘FILTER SECTION 


FIGURE 7-6. Single coil steam unit 


the same flow rate of outdoor air is introduced through 
properly designed replacement air heaters, the overall fuel 
cost does not exceed previous levels and often is decreased. 
A partial explanation of this savings is more efficient heat 
transfer. The most important factor, however, is that a well- 
designed air supply system is not dependent on the plant space 
heating system; rather, the two systems operate in an inde- 
pendent fashion. The air supply system and the plant heating 
system can be understood best by considering the building as 
a whole. In order for an equilibrium to be established, the heat 
outflow from the building must balance the heat inflow. To 
obtain additional energy saving during downtime, design the 
supply system to provide sufficient heating to counter air enter- 
ing the building through infiltration and to prevent freezing. 


7.11 REPLACEMENT AIR HEATING EQUIPMENT 


Replacement air heaters are usually designed to supply 
100% outdoor air. The basic requirements for an air heater are 
that it be capable of continuous operation, constant delivered 
air flow rate, and constant preselected discharge temperature. 
The heater must meet these requirements under varying con- 
ditions of service and accommodate outdoor air temperatures 
which vary as much as 40 F daily. Standard design heating 
and ventilating units are usually selected for mixed air appli- 
cations, 1.e., partial outdoor air and partial recirculated air; it 
is rare that their construction and operating capabilities will 
meet the requirements of industry. Such units are applicable 
in commercial buildings and institutional facilities where the 
requirements are less severe and where mixed air service is 
more common. 


Air heaters are usually categorized according to the source 
of heat: steam and hot water units, indirect-fired gas and oil 
units, and direct-fired natural gas and Liquified Petroleum 
Gas (LPG) units. Each basic type is capable of meeting the 
first two requirements — constant operation and constant 
delivered air flow rate. Variations occur within each type in 
relation to the third requirement, that of constant preselected 
discharge temperature. One exception to this rule is the direct- 
fired air heater where the inherent design provides a wide 
range of temperature control. Each type of air heater has 
specific advantages and limitations which must be understood 
by the designer in making a selection. 


Steam coil units were probably the earliest air heaters 


Replacement and Recirculated Air 7-9 


applied to general industry as well as commercial and institu- 
tional buildings (see Figure 7-6). When properly designed, 
selected, and installed, they are reliable and safe. They require 
a reliable source of clean steam at dependable pressure. For 
this reason they are applied most widely in large installations; 
smaller industrial plants often do not provide a boiler or steam 
capacity for operating a steam air heater. Principal disadvan- 
tages of steam units are potential damage from freezing or 
water hammer in the coils, the complexity of controls when 
close temperature limits must be maintained, high cost, and 
excessive piping. 


Freezing and water hammer are the result of poor selection 
and installation and can be minimized through careful appli- 
cation. The coil must be sized to provide desired heat output 
at the available steam pressure and flow. The coil preferably 
should be of the steam distributing type with vertical tubes. 
The traps and return piping must be sized for the maximum 
condensate flow at minimum steam pressure plus a safety 
factor. Atmospheric vents must be provided to minimize the 
danger of a vacuum in the coil which would hold up the 
condensate. Finally, the condensate must never be lifted by 
steam pressure. The majority of freeze-up and water hammer 
problems relate to the steam modulating type of unit which 
relies on throttling of the steam supply to achieve temperature 
control. When throttling occurs, a vacuum can be created in 
the coil and unless adequate venting is provided, condensate 
will not drain and can freeze rapidly under the influence of 
cold outdoor air. Most freeze-ups occur when outdoor air is 
in the range of 20-30 F and the steam control valve is partially 
closed, rather than when the outdoor air is a minimum tem- 
perature and full steam supply is on (see Figure 7-7). 


"Safety" controls are often used to detect imminent danger 
from freeze-up. A thermostat in the condensate line or an 
extended bulb thermostat on the downstream side of the coil 
can be connected into the control circuit to shut the unit down 
when the temperature falls below a safe point. As an alternate, 
the thermostat can call for full steam flow to the coil with 
shutdown if a safe temperature is not maintained. An obvious 
disadvantage is that the plant air supply is reduced; if the 
building should be subjected to an appreciable negative pres- 
sure, unit freeze-up still may occur due to cold air leakage 
through the fresh air dampers. 


The throttling range of a single coil unit can be extended 
by using two valves: one valve is usually sized for about 
two-thirds the capacity and the other valve one-third. Through 
suitable control arrangements both valves will provide 100% 
steam flow when fully opened and various combinations will 
provide a wide range of temperature control. Controls are 
complex in this type of unit and care must be taken to insure 
that pressure drop through the two valve circuits is essentially 
equal so as to provide expected steam flow. 


Multiple coil steam units (Figure 7-8) and bypass designs 
(Figure 7-9) are available to extend the temperature control 


7-10 


Industrial Ventilation 


STEAM COIL 


STEAM SUPPLY 
PROVIDE STEAM FROM A CLEAN SOURCE 
MAINTAIN CONSTANT PRESSURE WITH REDUCING VALVES IF REQUIRED 


PROVIDE TRAPPED DRIPS FOR SUPPLY LINES 
SIZE SUPPLY PIPING FOR FULL LOAD AT AVAILABLE PRESSURE 


STRAI 
DIAMETER MINIMUM PERFORATIONS 


ERTED BUCKET TRAP PREFERRED 
CONTROL VALVE 
SIZE FOR MAXIMUM STEAM FLOW 
AXIMUM PRESSURE DROP EQUAL TO 50% INLET STEAM PRESSURE 
SUUM BREAKER 
/2° CHECK VALVE TO ATMOSPHERE 
E VACUUM BREAKER 
ZE FOR DESIGN CAPACITY AT INLET STEAM PRESSURE (SUPPLY-—VALVE DROP) 
ERTICAL COILS PREFFERED 
HORIZONTAL COILS MUST BE PITCHED 1/4” PER FOOT TOWARD DRA\N. 
AXIMUM LENGTH RECOMMEMDED 
CONDENSATE TRAP 


A. INVERTED BUCKET PREFERRED 
B. SIZE TRAP FOR THREE TIMES MAXIMUM CONDENSATE LOAD AT PRESSURE 


DROP EQUAL TO 50% INLET PRESSURE 
C. INDIVIDUAL TRAP FOR EACH COIL 
CONDENSATE RETURN 
ATMOSPHERIC DRAIN ONLY 


AMERICAN CONFERENCE | 
Th 1 PIPING 
OF GOVERNMENTAL | pe CON ANG 


INDUSTRIAL HYGIENISTS 7 ee 


REHEAT COIL 
‘ V PREHEAT COIL 


=e \ 


° fo) 
fo) o| = 
are ° ° 
a fo) Oo} <a 
| ° ° 
XW ° O| —~— 
- ° ° 


FILTER SECTION 


FIGURE 7-8. Multiple coil steam unit 


range and help minimize freeze-up. With multiple coil units, 
the first coil (preheat) is usually sized to raise the air tempera- 
ture from the design outdoor temperature to at least 40 F. The 
coil is controlled with an on-off valve which will be fully open 
whenever the outdoor temperature is below 40 F. The second 
(reheat) coil is designed to raise the air temperature from 40 
F to the desired discharge condition. Temperature control will 
be satisfactory for most outdoor conditions, but overheating 
can occur when the outdoor air temperature approaches 40 F 
(39 F + the rise through the preheat coil can give temperatures 
of 79-89 F entering the reheat). Refined temperature control 
can be accomplished by using a second preheat coil to split 
the preheat load. 


Bypass units incorporate dampers to direct the air flow. 
When maximum temperature rise is required, all air is directed 
through the coil. As the outdoor temperature rises, more and 
more air is diverted through the bypass section until finally 
all air is bypassed. Controls are relatively simple. The princi- 
pal disadvantage is that the bypass is not always sized for full 
air flow at the same pressure drop as through the coil, thus 
(depending on the damper position) the unit may deliver 


Replacement and Recirculated Air 7-11 


y STEAM COIL 


™ FACE DAMPER 


©) 


4 / 
BY-PASS DAMPERS — teu ter SECTION 


FIGURE 7-9. By-pass steam system 


differing air flow rates. Damper air flow characteristics are 
also a factor. An additional concern is that in some units the 
air coming through the bypass and entering the fan compart- 
ment may have a nonuniform flow and/or temperature char- 
acteristic which will affect the fan’s ability to deliver air. 


Another type of bypass design, called integral face and 
bypass (Figure 7-10), features alternating sections of coil and 
bypass. This design promotes more uniform mixing of the air 
stream, minimizes any nonuniform flow effect and, through 
carefully engineered damper design, permits minimum tem- 
perature pickup even at full steam flow and full bypass. 


Hot water is an acceptable heating medium for air heaters. 
As with steam, there must be a dependable source of water at 
predetermined temperatures for accurate sizing of the coil. 
Hot water units are less susceptible to freezing than steam 
because of the forced convection which insures that the cooler 
water can be positively removed from the coil. Practical 
difficulties and pumping requirements thus far have limited 


OES 
reereeeees 


© 
Se 


OS 
SKS 
eR 


XS 


-, 
< 
x 


xo 
x 


ne 
KOR 


Q 


O} LE] SSSLLLLLI 
5 EE) (BI ERR 
Kole! Ke 
RRR RNY 
FEE SERINE 
CTE Kel > 
———— [0] £0) 
(ol (a 
+50° a 
Ca & 
——= (0) © 
“O) (5) 
c CG] fo] 
——= [6] ©) 
( (Ol fl 
(es 


FIGURE 7-10. Integral face and by-pass coil”) 


7-12 Industrial Ventilation 


COMBUSTION 
[ VENT FAN 


COMBUSTION fs! 


CHAMBERS: 


FILTER SECTION 


INLET i 


\ 


ZO 


FIGURE 7-11. \ndirect-fired unit 


the application of hot water to relatively small systems: for a 
100 F air temperature rise and an allowable 100 F water 
temperature drop, | GPM of water will provide heat for only 
450 cfm of air. This range can be extended with high tempera- 
ture hot water systems. 


Hybrid systems using an intermediate heat exchange fluid, 
such as ethylene glycol, have also been installed by industries 
with critical air supply problems and a desire to eliminate all 
freeze-up dangers. A primary steam system provides the 
necessary heat to a converter which supplies a secondary 
closed loop of the selected heat exchange fluid. The added 
equipment cost is at least partially offset by the less complex 
control system. 


Indirect-fired gas and oil units (Figure 7-11) are widely 
applied in small industrial and commercial applications. Eco- 
nomics appear to favor their use up to approximately 10,000 
cfm; above this size the capital cost of direct-fired air heaters 
is lower. Indirect-fired heaters incorporate a heat exchanger, 
commonly stainless steel, which effectively separates the 
incoming air stream from the products of combustion of the 
fuel being burned. Positive venting of combustion products is 
usually accomplished with induced draft fans. These precau- 
tions are taken to minimize interior corrosion damage from 
condensation in the heat exchanger due to the chilling effect 
of the incoming cold air stream. The indirect-fired air heater 
permits the use of oil as a heat source and room air recircula- 
tion is permitted with this type of unit since the air stream is 
separated from the products of combustion. A third major 
advantage is that this type of unit is economical in the smaller 
sizes and is widely applied as a "package" unit in small 
installations such as commercial kitchens and laundries. 


Temperature control, "turn-down ratio," is limited to about 
3:1 or 5:1 due to burner design limitations and the necessity 
to maintain minimum temperatures in the heat exchanger and 
flues. Temperature contro! can be extended through the use 
of a bypass system similar to that described for single coil 
steam air heaters. Bypass units of this design offer the same 


/ FRESH AIR 
LOUVERS 
RETURN AIR 
LOUVERS 


advantages and disadvantages as the steam bypass units. 


Another type of indirect-fired unit incorporates a rotating 
heat exchanger. Temperature control can be as high as 20:1. 


Direct-fired heaters wherein the fuel, natural or LPG gas, 
is burned directly in the air stream and the products of com- 
bustion are released in the air supply have been commercially 
available for some years (Figure 7-12). These units are eco- 
nomical to operate since all of the net heating value of the fuel 
is available to raise the temperature of the air resulting in a 
net heating efficiency approaching 90%. Commercially avail- 
able burner designs provide turndown ratios from approxi- 
mately 25:1 to as high as 45:1 and permit excellent 
temperature control. In sizes above 10,000 cfm, the units are 
relatively inexpensive on a cost per cfm basis; below this 
capacity, the costs of the additional combustion and safety 
controls weigh heavily against this design. A further disad- 
vantage is that governmental codes often prohibit the recircu- 
lation of room air across the burner. Controls on these units 
are designed to provide a positive proof of air flow before the 
burner can ignite, a timed pre-ignition purge to insure that any 
leakage gases will be removed from the housing, and con- 
stantly supervised flame operation which includes both flame 
controls and high temperature limits. 


Concerns are often expressed with respect to potentially 
toxic concentrations of carbon monoxide, oxides of nitrogen, 
aldehydes, and other contaminants produced by combustion 
and released into the supply air stream. Practical field evalu- 


PROFILE 
NY —-BURNER 


Leiter SECTION 


FIGURE 7-12. Direct-fired unit 


ations and detailed studies show that with a properly operated, 
adequately maintained unit, carbon monoxide concentrations 
will not be expected to exceed 5 ppm and that oxides of 
nitrogen and aldehydes are well within acceptable limits.) 


A variation of this unit, known as a bypass design, has 
gained acceptance in larger plants where there is a desire to 
circulate large air flows at all times (see Figure 7-13). In this 
design, controls are arranged to reduce the flow of outdoor air 
across the burner and permit the entry of room air into the fan 
compartment. In this way, the fan air flow rate remains 
constant and circulation in the space is maintained. It is 
important to note that the bypass air does not cross the burner 
— 100% outdoor air only is allowed to pass through the 
combustion zone. Controls are arranged to regulate outdoor 
air flow and also to insure that burner profile velocity remains 
within the limits specified by the burner manufacturer, usually 
in the range of 2,000 to 3,000 fpm. This is accomplished by 
providing a variable profile which changes area as the damper 
positions change. 


Inasmuch as there are advantages and disadvantages to both 
direct-fired and indirect-fired replacement air heaters, a care- 
ful consideration of characteristics of each heater should be 
made. A comparison of the heaters is given in Table 7-4. 


7.12 COST OF HEATING REPLACEMENT AIR 
As noted above, the cost of heating replacement air is 


TABLE 7-4. Comparison of Heater Advantages and Disadvantages 


Replacement and Recirculated Air 
Y ADJUSTABLE PROFILE DAMPER 


[AT 


= “ 


RECIRCULATED AIR LS ADJUSTABLE RECIRCULATING DAMPERS 


FIGURE 7-13. Direct fired by-pass unit 


probably the most significant annual cost of a ventilation 
system. Newer processes requiring cooling during the process 
must also be evaluated. Occupant comfort is more important 
than saving a few dollars in energy costs. Recent indoor air 
quality studies quantify diminished productivity when work- 
ers are uncomfortable. In addition to the equipment first cost, 
local building codes, and environmental regulations, designer 
experience in utility incentives and operating costs are in- 
volved in purchasing decisions. 


The American Society of Heating, Refrigerating, and 
Air Conditioning Engineers (ASHRAE), the U. S. De- 
partment of Energy, and others develop formulae and 
computer programs to determine the life-cycle costs of 
various equipment. The formulae and programs should 
not be used to determine the annual utility bill. Instead, 


Advantages Disadvantages 
Direct-fired Unvented: 


1. Good turndown ration -8:1 in small sizes; 25:1 in large sizes. Better 
control; lower operating costs. 


2. No vent stack, flue, or chimney necessary. Can be located inside 
walls of building. 


3. Higher efficiency (90%). Lower operating costs. (Efficiency based on 
available sensible heat.) 


4. Can heat air over a wide temperature range. 
5. First cost lower in large size units. 


Indirect Exchanger: 

1. No products of combustion; outdoor air only is discharged into 
building. 

2. Allowable in all types of applications and buildings if provided with 
proper safety controls. 

3. Small quantities of chlorinated hydrocarbons will not normally break 
down on exchanger to form toxic products in heated air. 


4, Can be used with oil, LPG, and natural gas as fuel. 
5. First cost lower in small size units. 
6. Can be used for recirculation as well as replacement. 


1. Products of combustion in heater air stream (some COs, CO, oxides 
of nitrogen, and water vapor present). 

2. First cost higher in small size units. 

3, May be limited in application by governmental regulations. Consult 
local ordinances. 


4. Extreme care must be exercised to prevent minute quantities of 
chlorinated or other hydrocarbons from entering air intake, or toxic 
products may be produced in heated air. 


5. Can be used only with natural gas or LPG. 


6. Burner must be tested to assure low CO and oxides of nitrogen 
content in air stream. 


1. First cost higher in large size units. 
2. Turn down ratio is limited — 3:1 usual, maximum 5:1. 


3. Flue or chimney required. Can be located only where flue or chimney 
is available. 


4. Low efficiency (80%). Higher operating cost. 

5. Can heat air over a limited range of temperatures. 

6. Heat exchanger subject to severe corrosion condition. Needs to be 
checked periodically for leaks after a period of use. 

7. Difficult to adapt to all combustion air from outdoors unless roof or 
outdoor mounted. 


7-14 Industrial Ventilation 


they are useful tools to compare the costs of various 
options in providing heating and cooling to an industrial 
ventilation system. 


The following two equations may be used to estimate replace- 
ment air heating costs on an hourly and yearly basis. They are 
based on average usage schedules and typical weather conditions 
rather than worst case conditions and maximum usage. 


Since there is an allowance for the efficiency of the replace- 
ment air unit, these equations will tend to give a low result if 
air is allowed to enter by infiltration only. They are also based 
on normal temperatures and moisture ratios and standard 
atmospheric pressure of 14.7 (101.4 kPa). Due to the heavy 
nature of the work in many industrial facilities, supply air may 
be cooler than for an office setting. Table 7-5 gives equation 
values (N) for supply air delivered at 70 F and 65 F. The 
humidity ratio (W) is assumed to be 0.01 pounds of water per 
pound of dry air. 


C, = Hourly cost =0.001 sola) c [7.1] 
q 
C. =Yearly cost = 0.154 (Q)(dg) (1) (¢) [7.2] 
q 


where: 
Q = air flow rate, cfm 
= required heat, BTU/hr/1000 cfm (Table 7-5 and 
Table 7-7) 
T = operating time, hours/week 
= available heat per unit of fuel (Table 7-6) 
dg = annual degree days (Table 7-7) 
c= cost of fuel, $/unit 


EXAMPLE PROBLEM 1 


Find the hourly and yearly cost of tempering 10,000 cfm 
of replacement air to 70 F in St. Louis, Missouri, using oil at 
$1.35/gallon. 


Average winter temperature = 31 F 


Hourly cost = = Cc 
=0.001x104 x 22000, $135 = $5.32 
106,500 
4 
Yearly cost = (0.154) (107) (6023) (40) x $135 


106,500 


= $4,700 (assuming 40 hrAweek 
of operation) 


The yearly cost is more representative because both the 
length and severity of the heating season are taken into 
account. 


TABLE 7-5. Required Heat for Outside Air Temperatures 


N, Required Heat, 
BTU/hr/1,000 cfm 


N, Required Heat, 


Avg. Outside Air BTU/hr/1,000 cfm 


Temperature, F @70F @ 65F 

0 77,000 71,500 

5 71,500 66,000 
10 66,000 60,500 
15 60,500 55,000 
20 55,000 49,500 
25 49,500 44,000 
30 44,000 38,500 
35 38,500 33,000 
40 33,000 27,500 
45 27,500 22,000 
50 22,000 16,500 
55 16,500 11,000 
60 11,000 5,500 
65 5,500 — 


NOTE: Sensible Heat Equation used: q = 1.1 (cfm) delta t. Humidity ratio is 
assumed to be 0.01 pounds of moisture per pound of dry air, 


7.13 AIR CONSERVATION 


The supply and exhaust of air represent both a capital cost 
for equipment and an operating cost which is often sizable in 
northern climates. Concerned designers, recognizing these 
cost and energy conservation needs, are unanimous in their 
desire for reduced ventilation rates. 


There are four methods by which the cost of heating and 
cooling a large flow of outdoor air can be reduced: 1) reduc- 
tion in the total flow of air handled, 2) delivery of untempered 
outdoor air to the space, 3) recovery of energy from the 
exhaust air, and 4) recovery of warm, uncontaminated air 
from processes. The successful application of these engineer- 
ing methods without reduction in health hazard control and 
without impairing the inplant environment requires careful 
consideration. 


7.13.1 Reduced Flow Rate: A reduction of total air flow 


TABLE 7-6. Available Heat per Unit of Fuel 


Available Btu 
Fuel Btu Per Unit Efficiency % Per Unit 
Coal 12,000 Btu/lb 50 6,000 
142,000 Btu/gal 

Oil 5 106,500 
Gas 

Heat 

Exchanger 1,000 Bult? 80 800 


Direct Fired 90 900 


Replacement and Recirculated Air 7-15 


TABLE 7-7. Heating Degree Day Normals and Average Winter Temperatures 


Phila- Pitts- Wash., 

City Albany Boston Chicago Cleveland Detroit Minneapolis NY delphia burgh St.Louis DC 

Avg Temp (F) 

Dec-~Feb 24 22.4 25 28 25.9 16 33.2 33.3 29 32.2 33.4 

Discharge Air 

Temp (F) Heating Degree Days 
80 11782 10409 10613 11343 10959 13176 9284 9652 10797 8943 8422 
79 11425 10049 10277 10982 10605 12826 8937 9300 10436 8624 8089 
78 11062 9690 9940 10621 10256 12478 8596 8954 10076 8310 7764 
77 10709 9242 9610 10265 9914 12135 8265 8619 9726 8003 7446 
76 10356 8994 9283 9915 9581 11797 7938 8285 9379 7702 7139 
15 10009 8652 8972 9570 9247 11475 7620 7959 9036 7443 6835 
74 9669 8317 8656 9229 8920 11142 7308 7641 8702 7121 6538 
73 9333 7790 8349 8898 8599 10816 7004 = 7328 8372 6839 6250 
72 9007 7668 8046 8567 8291 10496 6706 ©7028 8050 6560 5974 
1 8682 7354 7750 8248 7981 10180 6421 6728 7740 6289 5703 
70 8364 7046 7468 7928 7678 9870 6146 6438 7429 6023 5438 
69 8256 6749 7183 7617 7383 9567 5871. 6158 7127 5767 5179 
68 7750 6458 6905 7313 7100 9269 5606 5886 6833 5523 4929 
67 7452 6175 6635 7016 6816 8975 5349-5618 6546 5277 4690 
66 7162 5903 6373 6722 6543 8687 5101 5360 6272 5053 4455 
65 6881 5633 6122 6445 6278 8410 4858 5109 5997 4822 4229 
64 6607 5370 5875 6165 6020 8131 4621 4864 5734 4595 4014 
63 6340 5118 5638 5897 5772 7858 4394 4628 5483 4379 3798 
62 3081 4873 5399 5636 5533 7590 4176 4397 5234 4168 3588 
61 5829 4643 5164 5381 5290 7339 3957 4172 5006 3963 3383 
60 5586 4399 4936 5140 5054 7086 3747 =. 3952 4769 3761 3182 


rate handled can be accomplished by conducting a careful 
inventory of all exhaust and supply systems in the plant. 
Determine which are necessary, which can be replaced with 
more efficient systems or hood designs, and which systems 
may have been rendered obsolete by changes. 


Numerous hood designs presented in Chapter 10 are in- 
tended specifically to provide for adequate contaminant cap- 
ture at reduced air flow rates. For instance, the use of 
horizontal sliding sash in the laboratory hood can provide a 
30% saving in exhaust air flow rate without impairing cap- 
ture velocity. The use of a tailored hood design, such as the 
evaporation hood shown in VS-35-40, provides good contami- 
nant capture with far lower exhaust flow rates than would be 
required for a typical laboratory bench hood. Low Volume-High 
Velocity hoods and systems such as those illustrated in VS- 


40-01 through VS-40-20 are used for many portable hand tool 
and fixed machining operations and can provide contaminant 
capture at far lower air handling requirements. 


Throughout industry there are many applications of win- 
dow exhaust fans and power roof exhausters to remove heat 
or nuisance contaminants which would be captured more 
readily at the source with lower air flow rates. Many roof 
exhausters, as noted earlier, have been installed initially to 
combat problems which were really caused by a lack of re- 
placement air. When air supply and balanced ventilation con- 
ditions are established, their use no longer may be necessary. 


Good design often can apply proven principles of local 
exhaust capture and control to reduce air flow rates with 
improved contaminant control. 


7-16 Industrial Ventilation 


7.13.2 Untempered Air Supply: In many industries util- 
izing hot processes, cold outdoor air is supplied untempered 
or moderately tempered to dissipate sensible heat loads on the 
workers and to provide effective temperature relief for work- 
ers exposed to radiant heat loads. The air required for large 
compressors, as well as for cooling tunnels in foundries, also 
can come directly from outside the plant and thus eliminate a 
load that is otherwise replaced with tempered air. 


7.13.3 Energy Recovery: Energy recovery from exhaust 
air can be considered in two aspects: 1) the use of heat 
exchange equipment to extract heat from the air stream before 
it is exhausted to the outside and 2) the return (recirculation) 
of cleaned air from industrial exhaust systems. Heat ex- 
changer application to industrial exhaust systems has been 
limited primarily by the ratio of installed cost to annual return. 


Heat Exchangers — Air-to-air heat exchangers have been 
used to reduce energy consumption. This is achieved by 
transferring waste energy from the exhaust to replacement air 
streams of a building or process. The methods and equipment 
used will depend on the characteristics of the air streams. 
Major categories of equipment include heat wheels, fixed 
plate exchangers, heat pipes, and run-around coils. 


A heat wheel is a revolving cylinder filled with an air 
permeable media. As the exhaust air passes through the media, 
heat is transferred to the media. Since the media rotates, the 
warm media transfers heat to the cooler replacement air. 
Special care is required to ensure that this transfer does not 
cause a transfer of contaminants. 


A fixed plate exchanger consists of intertwined tunnels of 
exhaust and replacement air separated by plates (or sometimes 
a combination of plates and fins). The warm exhaust air heats 
the plates which in turn heat the cool replacement air on the 
other side of the plate. This exchanger uses no transfer media 
other than the plate forming wall of the unit. 


A heat pipe, or thermo siphon, uses a pipe manifold with 
one end in the warm exhaust air stream and the other in the 
cool replacement air stream. The pipe contains a fluid which 
boils in the warm exhaust air stream extracting heat and 
condenses in the cool replacement air stream releasing heat. 
Thus the heat pipe operates in a closed loop evaporation/con- 
densation cycle. 


A run-around coil exchanger uses a pair of finned-tube 
coils. A fluid circulates through the coils extracting heat from 
the warm exhaust air releasing heat to the cool replacement 
air. An advantage of the run-around coil is that the exhaust 
and supply duct systems can be separated by a significant 
distance which results in a reduced potential for re-entry; 
usually less duct in the systems and usually less roof area 
occupied by the units. 


Several factors are important in the selection of the appro- 
priate heat exchanger. A partial list is as follows: 


1. The nature of the exhaust stream. A corrosive or dust 


laden stream may need to be precleaned. 


2. The need to isolate the contaminated exhaust stream 
from the clean replacement air stream. 


3. The temperature of the exhaust stream. Unless the hot 
air stream is well above the desired delivery tempera- 
ture of the replacement air stream and the exhaust air 
stream is at elevated temperatures whenever heat is 
demanded by the replacement air stream, additional 
heating capacity will be required. 


4. Space requirements. Space requirements for some heat 
exchangers can be very extensive, especially when the 
additional duct runs are considered. 


5. The nature of the air stream. Many exhaust air streams 
are corrosive or dirty and special construction materi- 
als may be required. 


6. The need for a by-pass. During failure mode or sum- 
mer conditions, a by-pass will be required. 


Recirculation of Air from Industrial Exhaust Systems: 
Where large amounts of air are exhausted from a room or 
building in order to remove particulates, gases, fumes, or 
vapors, an equivalent amount of fresh tempered replacement 
air must be supplied to the room. If the amount of replacement 
air is large, the cost of energy to condition the air can be very 
high. Recirculation of the exhaust air after thorough cleaning 
is one method that can reduce the amount of energy con- 
sumed. Acceptance of such recirculating systems will depend 
on the degree of health hazard associated with the particular 
contaminant being exhausted as well as other safety, technical 
and economic factors. A logic diagram listing the factors that 
must be evaluated is provided in Figure 7-14.07 


Essentially this diagram states that recirculation may be 
permitted if the following conditions are met: 


1. The chemical, physical, and toxicological charac- 
teristics of the chemical agents in the air stream to be 
recirculated must be identified and evaluated. Exhaust 
air containing chemical agents whose toxicity is un- 
known or for which there is no established safe expo- 
sure level should not be recirculated. 


2. All governmental regulations regarding recirculation 
must be reviewed to determine whether it is restricted 
or prohibited for the recirculation system under re- 
view. 


3. The effect of a recirculation system malfunction must 
be considered. Recirculation should not be attempted 
if a malfunction could result in exposure levels that 
would cause worker health problems. Substances 
which can cause permanent damage or significant 
physiological harm from a short overexposure shall 
not be recirculated. 


4. The availability ofa suitable air cleaner must be deter- 
mined. An air cleaning device capable of providing an 


I, INITIAL DECISION 


IDENTIFY 
RECIRCULATION 
SYSTEM 


EVALUATE 
CHEMICAL AGENT 


| | 


CARCINOGEN OR NON ~ CARCINOGEN 
LOW SAFE EXPOSURE SAFE EXPOSURE 
LEVEL LEVEL 


ee 


NO ASSESS RESULT 
OF FAILURE 


ACCEPTA8LE 


CLEANER /MONITOR 
AVAILABILITY AND 
SUITABILITY 


NON—ACCEP TABLE 


NO 


| 


AVAILABLE AND 
SUITABLE 


NOT AVAILABLE 
OR SUITABLE 


os a 


ll, DESIGN AND ASSESSMENT 


DESIGN AND 
ACCESS SYSTEM 


COST SUITABLE 


il. SYSTEM EVALUATION 


CONSTRUCT AND 
INSTALL SYSTEM 


I 
em EVALUATE SYSTE 
I 
| 
WILL NOT MEET SAFE WILL MEET SAFE 
EXPOSURE LEVEL EXPOSURE LEVEL 
—— 


It 
“e+ CORRECT DESIGN 


OPERATE SYSTEM | 


FIGURE 7-14. Recirculation decision logic 


effluent air stream contaminant concentration suffi- 
ciently low to achieve acceptable workplace concen- 
trations must be available. 


5. The effects ofminor contaminants should be reviewed. 
For example, welding fumes can be effectively re- 
moved from an air stream with a fabric filter; however 
if the welding process produces oxides of nitrogen, 
recirculation could cause a concentration of these 
gases to reach an unacceptable level. 


6. Recirculation systems must incorporate a monitoring 


Replacement and Recirculated Air 7-17 


system that provides an accurate warning or signal 
capable of initiating corrective action or process shut- 
down before harmful concentrations of the recircu- 
lated chemical agents build up in the workplace. 
Monitoring may be accomplished by a number of meth- 
ods and must be determined by the type and hazard of the 
substance. Examples include area monitoring, for nui- 
sance type substances and secondary high efficiency 
filter pressure drop and on-line monitors for more haz- 
ardous materials. 


While all system components are important, special con- 
sideration should be given to the monitor. The prime requi- 
sites are that the monitor be capable of sensing a system 
malfunction or failure and of providing a signal which will 
initiate an appropriate sequence of actions to assure that 
overexposure does not occur. The sophistication of the moni- 
toring system can vary widely. The type of monitor selected 
will depend on various parameters (i.e., location, nature of 
contaminant — including shape and size — and degree of 
automation). 


7.13.4 Selection of Monitors: The safe operation of a re- 
circulating system depends on the selection of the best moni- 
tor for a given system. Reference 7.7 describes four basic 
components of a complete monitoring system which includes 
signal transfer, detector/transducer, signal conditioner, and 
information processor. Figure 7-15 shows a schematic dia- 
gram of the system incorporating these four components. It is 
quite likely that commercially available monitors may not 
contain all of the above four components and may have to be 
custom engineered to the need. 


In addition, the contaminant must be collected from the air 
stream either as an extracted sample or in toto. If a sample is 
taken, it must be representative of the average conditions of 
the air stream. At normal duct velocities, turbulence assures 
perfect mixing so gas and vapor samples should be repre- 
sentative. For aerosols, however, the particle size discrimina- 
tion produced by the probe may bias the estimated 
concentration unless isokinetic conditions are achieved. 


The choice of detection methods depends on the measur- 
able chemical and physical properties of the contaminants in 
the air stream. Quantifying the collected contaminants is 
generally much easier for particulate aerosols than for gases, 
vapors or liquid aerosols. 


Particulates: Where the hazardous contaminant constitutes 
a large fraction of the total dust weights, filter samples may 
allow adequate estimation of concentration. Better, if the 
primary collector (e.g., bag filters, cartridge filters) allows 
very low penetration rates, it may be economical to use high 
efficiency filters as secondary filters. If the primary filter fails, 
the secondary filter not only will experience an easily meas- 
ured increase in pressure drop, but will filter the penetrating 
dust as well — earning this design the sobriquet, "safety 


7-18 industrial Ventilation 


CALIBRATION (cs oe INFORMATION 
SIGNAL ~7 |PROCESSOR | 
| 
; ALARM 
y 1 HUMAN 
L INTERPRETATION 


p= | TRANSFER TRANSDUCER 


SIGNAL 
COND 


INDICATOR / 
RECORDER 


SIGNAL 
INPUT 
RECIRCULATING 
EXHAUST 
SYSTEM 


FIGURE 7-15. Schematic diagram of recirculation monitoring system 


monitor" systems (see Figure 7-16). 


Non-particulates: Continuously detecting and quantifying 
vapor and gas samples reliably and accurately is a complex 
subject beyond the scope of this manual. 


Air Sampling Instruments for Evaluation of Atmospheric 
Contaminants, published by ACGIH, describes and evalu- 
ates different air monitoring devices. The monitor in a recir- 
culating system must be capable of reliably monitoring 
continuously and unattended for an extended period of time. 
It must also be able to quickly and accurately sense a change 
in system performance and provide an appropriate warning if 
a preselected safety level is reached. In order to function 
properly, monitors must be extremely reliable and properly 
maintained. Monitors should be designed so that potential 
malfunctions are limited in number and can be detected easily 
by following recommended procedures. Required mainte- 
nance should be simple, infrequent, and of short duration. 


7.14 EVALUATION OF EMPLOYEE EXPOSURE LEVELS 


Under equilibrium conditions, the following equations may 
be used to determine the concentration of a contaminant 
permitted in the recirculation return air stream: 


(1—n) (Ce ~KpCw) 
1-[(Ke) (1- n)] 


R [7.3] 
where: 
Cr = air cleaner discharge concentration after 
recirculation, mg/m? 
1 = fractional air cleaner efficiency 
C, = local exhaust duct concentration before recircu- 
lation, mg/m? 
Kp = coefficient which represents a fraction of the 
recirculated exhaust stream that is composed 


SIGNAL DETECTOR / | 


AUTO 
RESPONSE 


MANUAL 
RESPONSE 


of the recirculation return air (range 0 to 1.0) 
Cy = replacement air concentration, mg/m? 


Q 
Cg =G, (Co -Cm) (1-f)+(Co — Cy) f 
A 


+KgCr +(1-Kg)(Cyy) [7.4] 


where: 


Cg = 8-hr TWA worker breathing zone concentration 
after recirculation, mg/m? 


Q, = total ventilation air flow before recirculation 
Qa, = total ventilation air flow after recirculation 
Cg = general room concentration before recircula- 
tion, mg/m? 
f = coefficient which represents the fraction of time 
the worker spends at the work station 


Co = 8-hrTWA breathing zone concentration at work 
station before recirculation 


Kg = fraction of worker’s breathing zone air that is 
composed of recirculation return air (range 0 to 
1.0) 


The coefficients Kp, Kp, and f are dependent on the work 
station and the worker’s position in relation to the source of 
the recirculation return air and the worker’s position in rela- 
tion to the exhaust hood. The value of Kp can range from 0 to 
1.0 where 0 indicates no recirculation return air entering the 
hood and 1.0 indicates 100% recirculation air entering the 
hood. Similarly, the value of Kg can range from 0 to 1.0 where 
0 indicates there is no recirculation return air in the breathing 
zone and 1.0 indicates that the breathing zone air is 100% 
recirculated return air. The coefficient "f"' varies from 0 where 
the worker does not spend any time at the work station where 
the air is being recirculated to 1.0 where the worker spends 


DAMPER MOTOR 
_— a 
Z 


BY-PASS saa 


DAMPER 


— 
ws: 
FAN 
CONTROLS 
[ PRIMARY 
COLLECTOR 
i 
C) wn 


mag S 
a 

==, PRS. MANOMETER ACROSS 
(\ | | \—_/] FABRIC COLLECTOR 


FIGURE 7-16. Schematic of recirculation from air cleaning devices (particulates) 


100% time at the work station. 


In many cases it will be difficult to attempt quantification 
of the values required for solution of these equations for an 
operation not yet in existence. Estimates based on various 
published and other available data for the same or similar 
operations may be useful. The final system must be tested to 
demonstrate that it meets design specifications. 


An example of use of Equations 7.3 and 7.4 and the effect 
of the various parameters is as follows: 


Consider a system with 10,000 cfm total ventilation before 
recirculation (Qs) consisting of 5,000 cfm of general exhaust 
and 5,000 cfm of local exhaust. The local exhaust is recircu- 
lated resulting in 10,000 cfm after recirculation air flow consist- 
ing of 5,000 cfm recirculated and 5,000 cfm fresh air flow. 


Assume poor placement of the recirculation return (Kp and 
Kg = I) and that the worker spends all his time at the work 
station (f = 1); the air cleaner efficiency (n) = 0.95; exhaust 
duct concentration (C;;) = 500 ppm; general room concentra- 
tion (Cg) = 20 ppm; replacement air concentration (Cy) = 5 
ppm; work station (breathing zone) concentration before re- 
circulation (Co) =35 ppm; anda contaminant TLV of 50 ppm. 


Equation 7.3 gives recirculation air return concentration: 


_ (10.95) (500—1x5) 


1[(yi-095)— PP 


R 


Equation 7.4 gives the worker breathing zone concentra- 


apl= 


Replacement and Recirculated Air 7-19 


AUTOMATIC ACTIVATION 
OF BY-PASS ON 
HIGH PRESSURE 


EXHAUST OUTLET 


DIRECT RECIRCULATED AIR 
AWAY FROM WORKERS 


ULTRA HIGH EFFICIENCY 
FILTER 


PRESSURE SWITCH 


ALARM 


MANOMETER 


tion: 


Cy “= (Cg —Cy) (1-f)+(Co — Cy) f+ KgCp 
+(1-Ke Cy) 


104 
= (2) (20-5) (1-1) + (35-5) (1)+ (26.1) 


+(1-1) (5) 


= 56.1 ppm 


Obviously, 56.1 ppm exceeds the TLV of 50 ppm and there- 
fore is unacceptable. 


In order to achieve lower concentrations (Cy), the system 
configuration must be redesigned so that only 50% of the 
recirculation return air reaches the work station. Thus, Kp and 
Kg are reduced to 0.5. Substituting these new data in Equation 
7.4, the breathing zone concentration calculates as 45.3 ppm. 
This is lower than the TLV of 50 ppm and therefore accept- 
able. 


Several potential problems may exist in the design of 
recirculated air systems. Factors to be considered are: 


1. Recirculating systems should, whenever practicable, 
be designed to bypass to the outdoors, rather than to 


7-20 


loo} 


Industrial Ventilation 


recirculate, when weather conditions permit. If a sys- 
tem is intended to conserve heat in winter months and 
if adequate window and door openings permit suffi- 
cient replacement air when open, the system can dis- 
charge outdoors in warm weather. In other situations 
where the work space is conditioned or where me- 
chanically supplied replacement air is required at all 
times, such continuous bypass operation would not be 
attractive. 


. Wet collectors also act as humidifiers. Recirculation 


of humid air from such equipment can cause uncom- 
fortably high humidity and require auxiliary ventila- 
tion or some means must be used to prevent excess 
humidity. 


. The exit concentration of typical collectors can vary 


with time. Design data and testing programs should 
consider all operational time periods. 


. The layout and design of the recirculation duct should 


provide adequate mixing with other supply air and 
avoid uncomfortable drafts on workers or air currents 
which would upset the capture velocity of local ex- 
haust hoods. 


. A secondary air cleaning system, as described in the 


example on particulate recirculation, is preferable to a 
monitoring device because it is usually more reliable 
and requires a less sophisticated degree of mainte- 
nance. 


. Odors or nuisance value of contaminants should be 


considered as well as the official TLV values. In some 
areas, adequately cleaned recirculated air, provided by 
a system with safeguards, may be of better quality than 
the ambient outdoor air available for replacement air 


supply. 


. Routine testing, maintenance procedures, and records 


should be developed for recirculating systems. 


. Periodic testing of the workroom air should be pro- 


vided. 


. An appropriate sign shall be displayed in a prominent 


place reading as follows: 


CAUTION 


AIR CONTAINING HAZARDOUS SUBSTANCES 
IS BEING CLEANED TO A SAFE LEVEL IN THIS 
EQUIPMENT AND RETURNED TO THE BUILD- 
ING. SIGNALS OR ALARMS INDICATE MAL- 
FUNCTIONS AND MUST RECEIVE IMMEDIATE 
ATTENTION: STOP RECIRCULATION, DIS- 
CHARGE THE AIR OUTSIDE, OR STOP THE 
PROCESS IMMEDIATELY. 


REFERENCES: 


Tels 


7.2. 


73. 


TA. 


75. 


7.6. 


Teds 


7.8. 


7.9. 


American Industrial Hygiene Association: Heating 
and Cooling Man and Industry. AIHA, Akron, OH 
(1969). 


Hart and Cooley Manufacturing Co.: Bulletin E-6. 
Holland, MI. 


Hama, G.: How Safe Are Direct-Fired Makeup Units? 
Air Engineering, p. 22 (September 1962). 


National Fire Protection Association, | Batterymarch 
Park, P. O. Box 9101, Quincy, MA 02269-9101. 


American Society of Heating, Refrigeration and Air 
Conditioning Engineers: Heating Ventilating and Air 
Conditioning Guide. ASHRAE, Atlanta, GA (1963). 


Hughes, R.T.; Amendola, A.A.: Recirculating Exhaust 
Air: Guides, Design Parameters and Mathematical 
Modeling. Plant Engineering (March 18, 1982). 


National Institute for Occupational Safety and Health: 
The Recirculation of Industrial Exhaust Air — Sym- 
posium Proceedings. Pub. No. 78-141 Department of 
Health, Education and Welfare (NIOSH), Cincinnati, 
OH (1978). 


American Conference of Governmental Industrial Hy- 
gienists: Air Sampling Instruments for Evaluation of 
Atmospheric Contaminants, 8th Edition. ACGIH, 
Cincinnati, OH (1995). 


American Society of Heating, Ventilating and Air 
Conditioning Engineers: HVAC Application. 
ASHRAE, Atlanta, GA (1995). 


Chapter 8 


VENTILATION ASPECTS OF INDOOR AIR QUALITY 


8.1 INTRODUCTION ..............-...00. 8-2 8.4.4 Heating/Cooling Coils... 2... ...022.. 8-6 
8.2. DILUTION VENTILATION FOR INDOOR AIR 8.4.5: Fans? on 2: att See dak dad % 8-6 
QUALITY «2... ee ee eee 8-2 8.4.6 Humidifier/Dehumidifier .... 2.2... ~«. 8-6 
8.3. HVAC COMPONENTS AND SYSTEM TYPES .. 8-2 8.4.7. Supply Air Distribution 2... 0... 8-6 
8.3.1 Components. ... 0.2.5.2... 2000-4 8-2 8.4.8 Supply Air Diffuser... 2... 8-7 
8.3.2 TypesofSystems.............-..- 8-3 8.4.9 Return AirGrilles ............... 8-8 
8.4 HVAC COMPONENTS, FUNCTIONS AND PA10: RetuniAir: cle koe ete hk oe Gs 8-8 
ML PUNG ONE Re k geet e ee S's Sot 8.4.11 FanCoil Units. ......0......0.. 8-8 
Gisink , TOMRDD OU MAN et af hee apo  g itt 8-4 85 HVAC COMPONENT SURVEY OUTLINE . .. . . 8-8 
8.4.2 Dampers............2..00-0 000] 8-5 REFERENCES........... 2... 8-10 
84.3 AirCleaning ...............08. 8-5 
Fig. 8-1 Single Duct Constant Volume With Reheat .. 8-11 Fig. 8-10 Filter Efficiency Vs. Particle Size... 2... . 8-20 
Fig. 8-2 Single Duct Constant Volume With Bypass .. 8-12 Fig. 8-11 Dampers: Parallel and Opposed Blade .... . 8-21 
Fig. 8-3 Single Duct Variable Air Volume With Reheat 8-13 Fig. 8-12 Typical HVAC Air Filters 2.2... 0.00.0. 8-22 
Fig. 8-4 Single Duct Variable Air Volume With Induction 8-14 Fig. 8-13 Heating/Cooling Coils... ........... 8-23 
Fig. 8-5 Single Duct Variable Air Volume with Fan Fig. 8-14 Terminology for Centrifugal Fan Components . 8-24 
Powered Devices and Reheat... 2... ..... 8-15 Fig. 8-15 HVAC Humidifier/Dehumidifier 2.2... .. 8-25 
Fig. 8-6 Dual Duct Constant Air Volume ......... 8-16 Fig. 8-16 HVAC System Self-Contained and Equipment 
Fig. 8-7 Single Duct Constant Air Volume Multi-Zone . 8-17 ROOM 225, sok Se GMs goes ee sete. PAR aed fs 8-26 
Fig. 8-8 Air Coil Unit. 2. en ee oe ce ea ee Sb ee 8-18 Fig. 8-17 Air Supply Diffusers and Return Air Grilles . . 8-27 
Fig. 8-9 Zone Heat Pump System ............-. 8-19 Fig. 8-18 Typical Partitioned Office Air Pattern... .. 8-28 


8-2 Industrial Ventilation 


8.1 INTRODUCTION 


There are two ventilation aspects which are major causes 
of the complaints noted in the vast majority of reported 
problems from all parts of this and other countries. They are 
complaints of unsatisfactory indoor air quality (which may be 
due to the lack of sufficient outdoor air for dilution of "nor- 
mal" indoor airborne contaminants) and the failure to deliver 
supply air properly to the occupied zones. 


Indoor air quality is defined as the overall quality of the 
indoor air and includes biological, chemical, and comfort 
factors. This chapter is designed to familiarize the reader with 
heating, ventilation, and air conditioning (HVAC) systems 
used in office and similar spaces. The individual components 
of a typical HVAC system are defined, and the operation of 
the more common types of HVAC systems found are dis- 
cussed. 


8.2 DILUTION VENTILATION FOR INDOOR AIR QUALITY 


The oil shortage and the resulting energy crisis of the late 
1960s and early 1970s is considered by some as the most 
significant cause of the current indoor air quality concern. In 
the past, when energy costs were relatively low, the design of 
heating, ventilation and air conditioning (HVAC) systems for 
buildings included the infiltration of outside air through 
doors, windows, and other sources. Also, up to 25% out- 
door air was supplied by the system, in addition to the 
infiltration, for general ventilation purposes. The outdoor air 
had the effect of diluting the "normal" indoor contaminants 
to a very low level of concentration, which had little effect on 
the occupants. 


Since the energy crisis resulted in major increases in energy 
costs, an extensive effort was made to reduce the infiltration 
of outdoor air by constructing the building as airtight as 
possible. Outdoor air supplied by the HVAC system was 
reduced to a minimum and in some instances eliminated 
entirely. Airborne contaminants found in indoor environ- 
ments were present in extremely small quantities and had not 
been a health problem in the past due to the dilution effect of 
the outdoor air. New concepts of office design that utilize 
fabric partitions, particle board furniture, increased use of 
carpets, office copy machines, etc., have increased the poten- 
tial for indoor contaminants. As buildings became more en- 
ergy efficient, there was an increase in complaints of 
stuffiness, drowsiness, tiredness, eye irritation, throat irrita- 
tion, and stale air. 


Existing health standards are not usually violated by the 
low-level concentrations, and the only current legal require- 
ment for outdoor air is found in the building codes. The 
Uniform Building Code®” is the most widely accepted stand- 
ard for providing outdoor air. Section 605 states that 5 cfm of 
outdoor air per occupant shall be mechanically supplied to all 
parts of the building during occupancy. Carbon dioxide con- 
centrations from occupant respiration within a space are often 


used as an indicator of the quantity of outdoor air being 
supplied to that space. When the indoor air concentration 
reaches approximately 800-1000 ppm (excluding external 
combustion sources), complaints may escalate. As the carbon 
dioxide levels increase, the number of complaints will in- 
crease more rapidly. 


In 1989, the American Society of Heating, Refrigeration 
and Air Conditioning Engineers (ASHRAE) developed and 
adopted ASHRAE Standard 62-1989, "Ventilation for Ac- 
ceptable Indoor Air Quality."®» The standard recognized the 
health problems resulting from the changes in construction 
and HVAC methods. It is based on occupancy of spaces and 
provides the outdoor air requirement for that space. Require- 
ments for outdoor air for offices based on an occupancy of 
seven people per 1000 square feet is currently 20 cfm per 
person. This is based on a total occupancy, including tran- 
sients, and is in addition to the usual HVAC requirements. 
The standard is expected to satisfy the requirements for 80% 
or more of the occupants. 


Provision for delivery of the outdoor air for dilution of the 
normal indoor airborne contaminants in the occupied space is 
a major factor of indoor air quality considerations. It is obvi- 
ous that if the outdoor air included as part of the total supply 
air is not delivered to the occupied zone, the potential for 
unsatisfactory indoor air quality increases. Another important 
factor in the delivery of the air to the occupied zone is the 
location of the supply and return air grilles to avoid short-cir- 
cuiting. Ideally, the supply air diffusers and the air grilles are 
so located that a uniform flow of air through the space occurs 
to avoid both stagnant air and drafts. 


Temperature and humidity can play a role in how people 
perceive indoor environment. ASHRAE Standard 55-1992@4) 
provides guidance in design and maintenance of indoor ther- 
mal environments. ASHRAE recommends temperature 
ranges of 67 to 76 F in winter (heating season) and 72 to 81 
F in summer (cooling season). However, complaints may 
increase when temperatures rise above 74 F. Similarly, it is 
preferable to keep relative humidities above 20-30% during 
the heating season and below 60% during the cooling season. 
ASHRAE also suggests limits on air movement. The average 
air movement in an occupied space should not exceed 30 fpm 
in winter or 50 fpm in summer. 


8.3 HVAC COMPONENTS AND SYSTEM TYPES 


When considering the ventilation aspects of HVAC sys- 
tems, the type of system and its components should be re- 
viewed for potential sources or causes of complaints regarding 
indoor air quality. Detailed descriptions of the systems and 
components can be found in the Systems and Equipment 
volume of the ASHRAE Handbook.®* 


8.3.1 Components: The components that make up HVAC 
systems generally include the following: 


. HVAC System: HVAC system refers to the equipment 
and distribution system used for heating, ventilating, 
cooling, humidifying, dehumidifying, and cleansing 
air for a building or building zone for the purpose of 
comfort, safety, and health of the occupants. 


. Dampers: Dampers are devices of various types used 
to vary the volume of air passing through an outlet, 
inlet, or duct. 


. Outdoor Air (Fresh Air; Replacement Air; Compen- 
sating Air): Outdoor air used to replace all or part of 
the air in a building or building space. 


. Return Air: Return air is air that has been in the 
building for a period of time and is returned to the 
HVAC system. Varying percentages of return air are 
exhausted outdoors with the remaining air (recircu- 
lated air) mixed with outdoor air for conditioning and 
distribution. 


. Mixing Plenum: A mixing plenum is a chamber 
within an HVAC system where outdoor air is mixed 
with returned air. The mixed air, after cleaning and 
conditioning, comprises the supply air for the building. 


. Air Cleaners: Air cleaners are devices designed to 
remove atmospheric airborne impurities such as dusts, 
gases, vapors, fumes, and smoke. (Air cleaners include 
air washers, air filters, electrostatic precipitators and 
charcoal filters.) 


. Heating Coils: Heating coils are heat transfer devices 
which utilize hot water, steam, or electricity to heat the 


supply air. 


. Cooling Coils: Cooling coils are heat transfer devices 
which utilize chilled water or a refrigerant to cool the 


supply air. 


. Condensate Pan (Drip Tray; Defrost Pan): A vessel 
or tray under the cooling coil to receive water extracted 
from the supply air by condensation from the cooling 
coil. 


. Humidifier/Dehumidifier. Humidifier/dehumidifiers 
are devices to add/remove moisture to/from the supply 
air. 


. Fans (Supply and Return): Fans are devices for mov- 
ing ventilation air through the HVAC system. 


. Supply Air: Supply air is conditioned ventilation air 
delivered to zones within a building. 


. Control Zone: Control zone is a space or group of 
spaces within a building served by an HVAC system. 
Depending on the space requirements, the control zone 
may be designated as core or interior zone and/or 
perimeter zone. 


Ventilation Aspects of Indoor Air Quality 8-3 


N. Occupied Zone: The occupied zone is the region 
within an occupied space between 3 and 72 inches 
above the floor. 


O. Control Box (Variable Air Volume, Bypass, Dual 
Duct): Control boxes are devices to which the supply 
air may be delivered by the HVAC system prior to 
delivery to the supply diffuser. These boxes may in- 
clude means of controlling supply air temperature and 
volume to the diffuser or multiple diffusers within a 
HVAC zone. 


P. Supply Air Diffusers: Supply air diffusers are devices 
whose function is to deliver the supply air to the 
occupied zone and to provide a desired distribution 
pattern. The diffusers may be circular, square, rectan- 
gular, linear slots, louvered, fixed, adjustable, or a 
combination. 


Q. Return Air Grilles: Return air grilles may be louvered 
or perforated coverings for openings located in the 
sidewall, ceiling or floor of a zone through which the 
return air enters. The return air grilles may be directly 
connected to an open return air plenum or to a ducted 
return air system. 


R. Return Air Plenum: A return air plenum is the space 
usually located above the ceiling where the return air 
is collected from a zone prior to entering the return air 
system. 


S. Economizer: An economizer is a control system which 
reduces the heating and cooling load through the use 
of outdoor air for free cooling when the total heat of 
the return air exceeds the total heat of the outdoor air. 


8.3.2 Types of Systems: There are different types of 
HVAC systems: single-duct systems, dual-duct systems, 
multi-zone systems, and special systems. These systems may 
be considered basic and subject to variations that are neces- 
sary to meet specific requirements. The following descrip- 
tions of the basic systems are intended as a guide and the 
referenced ASHRAE Handbook should be reviewed for sys- 
tem details and variations. 


Single-Duct Systems may be either a constant or a variable 
air volume system. The constant volume system maintains 
constant air flow with the temperature of the supply air 
controlled in response to the space load. See Figures 8-1 and 
8-2. The system may be a single zone, a zoned reheat, multi- 
ple-zone modification or a by-pass variation using a by-pass 
box in lieu of reheat constant volume primary system with a 
variable air volume secondary system. A variable air volume 
(VAV) system controls the temperature within a zone by 
varying the supply air volume. See Figures 8-3, 8-4, and 8-5. 
This type of system may include reheat at the terminals, 
induction unit, fan-powered distribution box, dual conduit, 
and variable diffusers. 


8-4 Industrial Ventilation 


Dual-Duct Systems condition all the air in a central appa- 
ratus and distribute it to the conditioned zones through two 
parallel mains, one carrying cold air and the other warm air. 
The system may be a constant volume type single fan and with 
or without reheat capability. See Figure 8-6. Also, the system 
may be VAV which mixes the cold and warm air in various 
volume combinations depending on the zone load. In both 
system types, the cold and warm air is delivered to a dual duct 
box which mixes the air prior to delivery to the supply air 
diffuser. 


Mutltizone Systems supply several zones from a centrally 
located HVAC unit. Supply air for the different zones consists 
of mixed cold and warm air through zone dampers in the 
central HVAC unit in response to zone thermostat control. 
From there, the supply air is distributed through the building 
by single zone ducts which, in turn, supply the air to the zone 
diffusers. See Figure 8-7. 


Fan Coil Units are usually located along the outdoor wall 
of a building for heating and cooling the perimeter up to 15 
feet from the outdoor wall. These units may have a through- 
wall duct for outdoor air and can be totally self contained or 
have the heating and cooling media supplied from a central 
mechanical room. See Figure 8-8. Controls for temperature 
and operation will vary although control of the outdoor air is 
usually at the unit and beside the nearest occupant. 


Zone Heat Pumps are packaged HVAC units that may 
provide the heating and cooling for individual zones within a 
building. These units vary in how the heating and cooling 
media is provided, but the function is generally constant (see 
Figure 8-9). Also, these units may be located within the 
individual zone above the ceiling in the return air space or 
remotely such as on the building roof. The supply air is 
delivered to the entire zone through a duct distribution and 
diffuser system. Return air for a pump located in the building 
enters the return air plenum above the ceiling due to zone 
pressure and migrates to the unit for reconditioning. For the 
remote unit, the return air is ducted from the ceiling plenum 
or from return air grilles to the unit. Outdoor air for interior 
units may be provided by a separate system and delivered to 
the return air plenum above the ceiling. Some building codes 
require that the outdoor air be directly supplied to the interior 
units. For the remote unit located on the roof, the outdoor air 
may be provided by the unit on the return air side through a 
damper that usually is set manually. 


8.4 HVAC COMPONENTS, FUNCTIONS, AND 
MALFUNCTIONS 


8.4.1 Outdoor Air: The outdoor air requirement for a 
space or an entire building must satisfy the need for acceptable 
indoor air quality and the need to replace air removed from 
the space or building by process or other exhaust. For indoor 
environment, ASHRAE Standard 62-1989, "Ventilation for 
Acceptable Indoor Air Quality,"®» is the accepted design 


criteria. Replacement air, however, will depend on factors 
such as total exhaust volume and pressure differential require- 
ments of the space or building plus the evaluation of potential 
airborne contaminants that may be generated inside or outside 
the building. For example, in the "open concept" type of office 
layout where partitions approximately five feet high enclose 
office spaces, the supply air has a tendency to ventilate only 
the space between the partitions and the ceiling. Very little, if 
any, of the supply air enters the actual occupied space directly 
to provide the necessary dilution. This allows contaminants 
in the occupied space to increase in concentration resulting in 
the potential for unsatisfactory air quality complaints. 


ASHRAE Standard 62-1989 recommends the measure- 
ment and documentation of the outdoor air intake volumetric 
flow rate on all configurations of HVAC systems. The pri- 
mary purpose of this requirement is to contro! the level of 
carbon dioxide, human odors, and the normal airborne con- 
taminants generated within the space. The published ventila- 
tion rates are based on occupancy or space usage and on an 
assumed occupant density. If the occupant density increases 
or the space usage increases, a degradation of the indoor air 
quality will occur which, in turn, will require an increase in 
outdoor air. 


It is possible to estimate the percentage of outdoor air by 
equation using the return, outdoor, and mixed air tempera- 
tures. The percentage of outdoor air also can be determined 
by equation using the carbon dioxide concentrations in the 
same air flow areas. Using these results, the estimated volu- 
metric flow rate of the outdoor air can be determined. The 
equations are as follows: 


Temperature Method: 


° tea — 
% outdoor air = tra = tMa x100 
tra — toa 
where: 
tra = temperature, return air 
tua = temperature, mixed air 


toa = temperature, outdoor air 
Carbon Dioxide Method: 


ppm/RA-—ppm/MA 
ppm/RA-—ppm/ OA 


% outdoor air = x 100 


where: 
ppm/RA = CO2 concentration, return air 
ppm/MA = COQ2 concentration, mixed air 
ppm/OA = COz2 concentration, outdoor air 
Given the legal implications, the direct measurement and 


documentation of the outdoor air volumetric flow rate is 
recommended. 


TABLE 8-—1® Relationships among extent of complaints regarding 
indoor air quality, CO, levels, and outdoor air ventilation rates 


Outdoor Air 
Ventilation Rate/Person 

COMMENTS CO, (ppm) CFM Ls 
Occasional complaints, 600 35 16.5 
particularly if the air 
temperature rises 
Complaints are more 800 21 10 
prevalent 
Insufficient replacement air, 1000 15 7 


complaints more general 


The concentration of carbon dioxide within a space may 
provide a good indication of the outdoor air being delivered 
to the space. A study conducted by the Ontario Inter-Minis- 
terial Committee on Indoor Air Quality reported on the rela- 
tionship between levels of complaints, carbon dioxide 
concentrations, and the outdoor air ventilation rates. The 
results are indicated in Table 8-1. 


Location of outdoor air intake may be found on the building 
roof, sidewall, at ground level, or possibly at all three loca- 
tions for very large building complexes. Figure 5-28, "Air 
Flow Around Buildings," clearly illustrates the potential for 
airborne contaminants to enter the building through any of the 
outdoor air intakes. Sources of potential airborne contami- 
nants from the building and from sources remote or adjacent 
to the building should be thoroughly investigated. Assistance 
with this investigation should be requested of industrial hy- 
giene and environmental organizations who have responsi- 
bilities for the building. The location of the outdoor air intakes 
will be affected by the atmospheric air flow over the building 
as will the location and height of any exhaust stacks. Criteria 
for the atmospheric air flow characteristics and stack heights 
may be found in Figures 5-28, 5-29, and 5-30 of this manual 
or in the Fundamentals volume of the 1993 ASHRAE Hand- 
book, Chapter 14, "Airflow Around Buildings."@? 


Roof intakes generally are located within a few feet of the 
roof surface. Standing water on the roof from weather condi- 
tions, HVAC equipment drains, or other sources present the 
potential for biological growth and entry into the intake. 
Unless discharged vertically above the recirculation region 
(see Figure 5-28), building exhaust systems from restrooms, 
processes within the building, and restaurant kitchens have 
significant potential for re-entry. 


Building sidewall intakes have the potential for entry of 
airborne contaminants from street level automotive traffic, 
shipping and receiving docks, and adjacent buildings. In the 
building wake region, see Figure 5-28, the potential for re-en- 
try increases for the outdoor air intakes and open windows or 
doors since the pressure in the recirculation region is lower 


Ventilation Aspects of Indoor Air Quality 8-5 


than the surrounding area. Airborne debris such as leaves, 
paper, and atmospheric dirt may tend to collect on the intake 
bird screens which can reduce the intake area and may reduce 
the flow rate into the intakes. 


Ground level outdoor air intakes are possibly the least 
desirable location of the three described. This location offers 
the potential for air quality problems caused by standing 
water, automotive emissions, and as a collection point for dirt 
and debris. The security of a building can be compromised 
through the ground level intakes by the deliberate addition of 
foreign materials. 


8.4.2 Dampers: A typical office building HVAC system 
will include outdoor and return air dampers. Air flow through 
these dampers will vary over a wide range depending on the 
damper opening settings and the space or building require- 
ments. Indoor air quality problems often result if the outdoor 
air damper is not designed or adjusted to allow introduction 
of sufficient outdoor air for the current use of the building. 
Outdoor air requirements for acceptable indoor air quality 
indicate that the actual volumetric flow rate through the 
damper sections be monitored. When the outdoor air and 
return air dampers are combined in an HVAC system, there 
may be an imbalance in volumetric flow rates. 


Itis customary to report the volumetric air flow rate through 
dampers in terms of damper opening. However, damper open- 
ing is not linearly proportionate to volumetric flow rate. 
Another misconception regarding dampers is that a "closed" 
damper will leak approximately 10%. Closed dampers may 
not leak at all. 


Dampers are mechanical devices (either parallel or opposed 
blade, see Figure 8-11) that require routine maintenance and 
periodic settings checks to assure proper air flow passage. 
Actuators, connecting arms and damper bearings are compo- 
nents that can affect the air flow if not properly connected or 
adjusted. This is an area where potential problems affecting 
the indoor air quality are not uncommon. In older buildings, 
the practice of disconnecting the outdoor air dampers to 
conserve energy is fairly common. This practice has been 
found by current surveys in some older buildings and also in 
newer buildings which have been occupied for an extended 
period. The result, of course, is a significant potential source 
of unsatisfactory indoor air quality complaints. 


8.4.3 Air Cleaning: The requirements for air cleaning vary 
according to the space or building requirements. There are, 
however, some basic factors that should be included in the 
design of the HVAC filter section. It is considered essential 
that all ventilation supply air, including outdoor and recircu- 
lated air, pass through a prefilter and a high efficiency final 
filter. Depending on requirements, filters such as charcoal, 
potassium permanganate, HEPA, and others may be speci- 
fied. See Figure 8-12 for various types. 


8-6 Industrial Ventilation 


For example, paper dust is one of the contributors to 
unsatisfactory indoor air quality. The paper dust in itself is an 
irritant to the eyes and respiratory system. Also, many papers 
are chemically treated which tends to compound the irritant 
effect. The dust generated enters the return air section of the 
HVAC system, and may be reintroduced to the space being 
served by the system. Studies®**! have indicated that a 
significant percentage of the paper dust will be removed by high 
efficiency type filters which, as stated, should be included in the 
HVAC filter section. Figure 8-10(8.5) shows approximate effi- 
ciency versus particle size for typical air filters. 


In some older HVAC equipment, also in perimeter fan coil 
units and self-contained heat pumps, low efficiency filters are 
noted. It may be possible to replace these filters with medium 
efficiency filters of the same dimensions. The medium efficiency 
pleated filter has more than twice the filtering area, thereby 
increasing the interception of the airborne particulates without a 
significant increase in the static pressure requirements. 


The air cleaning or filter section of the HVAC system 
requires routine maintenance for replacing dirty filters, or in 
some instances, cleaning a reusable type of air cleaner. Rou- 
tine maintenance would also bring attention to damaged filters 
or filter frames and uneven air flow (by the dirt pattern on the 
face of the filters.) Even though a regular maintenance pro- 
gram may be in force, the filter sections, both pre- and final 
filters, present a potential source for indoor air quality com- 
plaints. 


Some HVAC systems utilize self-contained heat pumps to 
control conditions in specific zones. These heat pumps can be 
located above the ceiling in the return air plenum near the zone 
being served. This location results in a difficult situation in 
terms of providing service for the unit. The zonal heat pump 
usually has a low efficiency filter which can be completely 
blocked or missing due to the difficulty of servicing. 


8.4.4 Heating/Cooling Coils: Heating and cooling coils 
must be free of damage, especially the heat transfer fins. 
Irregularities in the fins will result in unequal heat transfer and 
will provide an area for dirt and other materials to accumulate. 
Air cleaning sections are not 100% perfect in removing the 
airborne contaminants regardless of efficiency ratings. Con- 
venient access to the coil section for inspection, cleaning and 
maintenance is essential to the proper functioning of the coil. 
See Figure 8-13. 


Cooling coils require some additional considerations. The 
supply air will pass through at a relatively low velocity, and 
the heat transfer will condense moisture on the coil fins. This 
moisture will drain to the condensate pan below the coil. 
Provision must be made to properly discharge the condensate. 
Since the air cleaning section is not perfect, some airborne 
contaminants will reach the cooling coils. The moisture accu- 
mulating on the coil fins will collect a significant percentage 
of these contaminants, which may adhere to the fins or drain 


to the pan below with the condensate. Accumulations of these 
contaminants create a source of molds, spores, bacteria, etc., 
that may enter the supply air stream. The condensate pan drain 
may allow condensate to accumulate at or near the outdoor 
air intake and can re-enter the HVAC system. It is essential 
that the cooling coil condensate pan be properly drained. The 
pan drain must be directed away from any outdoor air intake. 
The coil and pan must be inspected and cleaned on a reguiar 
basis. Microorganisms may proliferate if this is not done. 


8.4.5 Fans: HVAC systems vary in size and complexity 
over a wide range as do the fans as the system prime mover 
of volumetric flow for both supply, outdoor and return air. 
The fans may be the axial or centrifugal type with inlet vanes, 
outlet dampers, variable speed, direct or belt drive (see Figure 
8-14). Also, the fan or fans may be inside the housing of a 
self-contained HVAC unit or a separate component in a 
mechanical room or penthouse. See Figure 8-16 for a typical 
layout of the self-contained and mechanical room system. 


Since the fan is the prime mover of the HVAC system, a 
preventive maintenance program usually will reveal any po- 
tential malfunctions before they occur. Failures of the fan are 
usually noted immediately and corrected by maintenance. 
There is, however, a maintenance procedure (lubrication of 
moving components) that may be a source of odor complaints 
by the building occupants. Over lubrication, which is not an 
uncommon practice, may place a small quantity of the lubri- 
cant in the air flow into the fan. This may cause the blades to 
become coated and the lubricant odor to be carried into 
occupied areas. 


8.4.6 Humidifiers/Dehumidifiers: The incorporation of 
humidifiers/dehumidifiers is dependent on the space require- 
ments of the building. Humidifiers add moisture to the supply 
air by direct water or by steam spray. Dehumidifiers remove 
moisture from the supply air by a desiccant-type filter or by 
cooling coils. Of the two processes, humidification is more 
widely used in HVAC systems (see Figure 8-15). The equip- 
ment used for humidification has a reputation for requiring a 
high level of maintenance for proper operation. For this 
reason, it is fairly common to find that the humidifier has been 
shut off — especially in office buildings. 


Both humidifying and dehumidifying are associated with 
water and dampness. This association presents the potential 
for the growth of molds, spores, bacteria, etc., that may enter 
the supply air flow. Proper drainage of any water or moisture 
generated by either process is essential. 


8.4.7 Supply Air Distribution: The air supply distribution 
system should be through sheet metal, steel and aluminum, or 
some type of non-fibrous duct material. There is increasing 
concern that fibrous materials such as fiberglass board ducts 
may produce fibers that may be potentially harmful. The use 
of interior duct insulation should also be avoided to eliminate 
the possibility of fibers entering the supply air stream. Supply 


air duct systems should be designed in accordance with ac- 
cepted standards as detailed in current publications such as 
the ASHRAE Handbook series, standards of the Sheet Metal 
and Air Conditioning Contractors National Association 
(SMACNA), National Fire Protection Association, and other 
applicable criteria sources. 


The physical condition of the air supply duct system is 
important in the overall evaluation. Duct systems are usually 
located above the ceiling in the return air space together with 
utility lines, sprinkler lines, computer cables, etc. This space 
is relatively small, and when repairs, rearrangements, instal- 
lations, etc., occur, damage to the duct system may not be 
noticed but may affect the air distribution. 


It is common practice to connect the supply air duct to 
mixing boxes and/or diffusers with flexible duct. The fric- 
tional resistance can be up to five times that of sheet metal, 
and the manufacturer’s data should be reviewed. Bends and 
turns using flexible duct will compound the losses and have 
a tendency to reduce the cross-section, which may in turn 
reduce the air volume. Improper hangers and supports also 
have the same tendency and results. The use of flexible ducts 
should be limited to minimum lengths, properly supported 
and securely fastened at each end. 


8.4.8 Supply Air Diffuser: The function of the supply air 
diffuser is to deliver and distribute the supply air throughout 
the occupied zone. Diffusers are available in a wide variety 
of types, shapes, and sizes — all of which will provide the 
proper air volume according to the supplier. See Figure 8-17 
for illustrations of various types of diffusers. The suppliers or 
manufacturers usually rate the diffusers in terms of supply air 
volume, static pressure drop, and the "throw" or pattern of the 
air delivery. Also, the published data will include illustrations 
of the air flow pattern created by the diffuser in a totally empty 
space and rely on the “coanda effect" for mixing the air in the 
space. However, when the space is occupied by people, 
equipment, file cabinets, cubicle partitions, library shelves, 
etc., the supply air pattern changes dramatically. This results 
in less supply air to the occupied zone. This ventilation aspect 
can be easily recognized through the use ofa simple hand-held 
smoke test. 


Some of the more common problems associated with dif- 
fusers are as follows: 


1. Variable air volume HVAC systems with fixed-supply 
air diffusers vary the air flow rate depending on tem- 
perature demand. Even if the minimum outdoor air is 
provided at all times, the reduced flow rate through the 
diffusers will reduce the throw and flow pattern. This 
may result in some areas within the occupied zone 
receiving little or no supply air. There are diffusers that 
automatically adjust for reduced air flow rates to main- 
tain a constant throw and flow pattern utilizing the 
"coanda effect." Reports from the field vary over a wide 


Ventilation Aspects of Indoor Air Quality 8-7 


range. Some reports state that technical maintenance 
is relatively high to assure proper functioning. Others 
report that the pressure required increases as the slot 
area decreases which decreases the air flow rate. 


2. Variable air volume systems may include constant air 
flow to the diffusers through terminal boxes serving 
specific zones. These boxes contain an air supply fan 
with sensors and controls that draw air from the return 
air system or the specific zone based on the flow rate 
from the main system. Even though this will assure a 
constant air flow rate to the diffuser, it also presents 
some potential problems by localized recirculation 
within the specific zone. 


3. Supply air diffusers with fixed blades, diffusers cov- 
ered by a perforated plate, linear fixed diffusers, and 
fluorescent light troffers direct the supply air across 
the ceiling depending on the coanda effect for delivery 
to the occupied zone. In the "open concept" office 
layout with 5-ft high partitions enclosing office spaces, 
the supply air from the diffusers described will have a 
tendency to provide continuous supply only to the 
space between the partitions and the ceiling. See Fig- 
ure 8-18. Very little if any of the supply air enters the 
occupied zone directly which may result in com- 
plaints. This particular ventilation aspect may occur 
even though the system is providing 100% outdoor air 
and is often referred to as "short circuiting." The flow 
pattern above the partitions can usually be observed by 
using the simple smoke test. Another test that is done 
which may give a more qualitative result is the meas- 
urement of the carbon dioxide concentrations in the 
occupied zone and in the space between the partitions 
and ceiling. A concentration in the occupied zone that 
is significantly higher than the concentration above the 
partitions indicates that possibly up to 75% of the 
supply is above the partitions. 


4. Supply air diffusers with adjustable blades are avail- 
able in the multi-directional ceiling type, linear dif- 
fusers with adjustable T-bars, sidewall supply grilles, 
and other types. The adjustable feature does offer a 
means of better directing the supply air to the occupied 
zone. However, locations of the diffusers and the 
adjustment of the blades is critical to the distribution 
of the supply air. Improper adjustment may result in 
complaints by the occupants of excessive drafts. 


Location and type of supply air diffusers should be such 
that a continuous flow of air through the occupied space will 
occur at all times. Avoid situations that result in localized 
recirculation or short circuiting to the return air system. In 
general, the air flow pattern through a space by the supply air 
should receive critical attention and can be characterized in 
terms of ventilation efficiency. Two efficiencies should be 
considered: system efficiency and ventilation efficiency. Sys- 


8-8 Industrial Ventilation 


tem efficiency is defined as the ratio of the actual volumetric 
flow rate to a specific space to the design volumetric flow rate 
for that specific space. Ventilation efficiency is defined as the 
ratio of the actual volumetric flow rate to a specific occupied 
zone to the design volumetric flow rate for that specific 
occupied space. Location and type of supply air diffusers are 
critical in the development of good ventilation efficiency. 
Design criteria in ASHRAE Standard 62-1989 will assist the 
design engineer in this effort. 


8.4.9 Return Air Grilles: The return grilles have the func- 
tion of receiving or exhausting air from a space through the 
return air system. Also, it is the function of the return air grilles 
to enhance the flow of the supply air through the space. The 
size and number of return grilles must be such that 100% of 
the supply air can be returned to the return air system. Loca- 
tion of the return air grilles influences the air flow pattern 
through the space and proper location will minimize localized 
recirculation zones. 


There is little design data available on the placement of the 
return air grilles but the location should be considered as 
important as the location of the supply air diffusers. The 
short-circuiting of the supply air directly to the ceiling return 
grilles may result in less than 50% of the supply air reaching 
the occupied zone. Development of an air flow pattern 
through an occupied zone from the supply diffusers to the 
return air grilles is a primary consideration. 


8.4.10 Return Air: The return air system may be either an 
open plenum type or a ducted system, both of which are 
typically located above the ceiling. In the return system, a 
static pressure balance between return air points must be part 
of the system design. It is obvious that the open ceiling plenum 
cannot be balanced by design which accounts for difficulty in 
providing a balanced supply air volume. For ducted return, 
the approach is similar to an industrial exhaust system. The 
static pressure in each run must be balanced by design at their 
junction which also accounts for difficulty in providing a 
balanced supply air volume. 


Pressure differentials at any junction are limited to 20% 
which is the maximum correction possible by damper. For 
differentials over this limit, redesign is necessary. Flexible 
duct is used at times to connect return air grilles to the ducted 
return. Since the negative pressure will tend to collapse the 
flexible duct, this practice should be avoided. 


8.4.11 Fan Coil Unit: The fan coil units used for HVAC 
are commonly located around the perimeter of a building and 
serve up to 15 feet from the outdoor wall. See Figure 8-8 for 
an illustration of a typical fan coil unit. These units may be 
totally self-contained with automatic controls; may include a 


through-wall duct for outdoor air; may have remote heating 
and cooling media or may be controlled manually at the unit. 
Since the fan coil units are rather compact, the filters are 
relatively small and in the low efficiency range. This will tend 
to increase the maintenance requirements since the return air 
is at the floor level — a potential significant source for dirt 
and possibly other contaminants. A provision for outdoor air 
may be a feature of the fan coil unit especially for units used 
to provide the HVAC for the building perimeter. The outdoor 
air intake is normally screened and may, over time, become 
blocked by dirt and debris from the outside atmosphere. Also, 
the intake may be located on an outside ledge of the building, 
depending on the building design, which may be a roosting 
area for birds. The outdoor air intake presents a significant 
source for contaminants and a difficult location to maintain, 


8.5 HVAC COMPONENT SURVEY OUTLINE 


The responsibility for monitoring the indoor air quality 
within a building may be assigned to an office individual, the 
building maintenance department, an outside environmental 
firm, or an HVAC maintenance contractor. In order to meet 
this responsibility, the assignee should conduct periodic walk- 
through surveys of the HVAC system and its components. The 
assignee should have a procedure or outline of the system compo- 
nents in order to conduct the survey. Basic information required to 
develop the procedure would including the following: 


1. The mechanical plans and specifications for the 
HVAC system to be surveyed including modifications 
or rearrangements, which are essential to conducting 
the survey. 


2. A detailed description of the type of HVAC system, 
its features and functions, especially for those who are 
not thoroughly acquainted with the system. 


3. The current test and balance reports which can provide 
information on air distribution and design vs. perform- 
ance data. These reports may also indicate a specific 
component problem such as outdoor air requirements. 


4. Reports of complaints regarding the indoor air quality 
(which should include the nature and location). These 
reports are essential to conducting this survey. They 
may indicate a component problem such as a discon- 
nected diffuser and the lack of air movement in an 
occupied zone. 


In addition to the basic information, the walk-through 
survey includes observation or inspection of each of the 
HVAC system components for potential malfunction. The 
procedure or survey outline of the components together with 
specific notes follows. 


Ventilation Aspects of Indoor Air Quality 8-9 


WALK-THROUGH SURVEY OUTLINE 


1. Outdoor Air (see Figure 5-28): 


A. Intake location and physical condition 


B. Building exhaust stacks and vent pipes adjacent to 
intake 


C. Cooling tower; type and location 


D. Building entryways, doors, and windows as poten- 
tial entries for airborne contaminants 


E. Areas adjacent to the building as potential sources: 
shipping/receiving docks, parking lots, high traffic 
roads, adjacent buildings and operations, etc. 


2. Dampers (see Figure 8-11): 
A. Outdoor air; type and physical condition 
B. Return air; type and physical condition 
C. Face and bypass; type and physical condition 


D. Exhaust/pressure relief; type and physical condition 


3. Air Cleaning (see Figure 8-12): 


A. Type and general condition 
B. Prefilter; type, efficiency, and condition 


C. Final filter; type, efficiency, and condition 


4. Heating/Cooling Coils (see Figure 8-13): 


A. Pre-heat; type and condition 
B. Cooling; type and condition 
C. Condensate pan and drain 


D. Re-heat; type and condition 


5. Fans/Blowers (see Figure 8-14): 


A. Supply air; type and condition 


B. Return air; type and condition 


C. Exhaust/pressure relief; type and condition 


6. Humidifier/Dehumidifier (see Figure 8-15): 


A. Type and general condition 


B. Condensate pan and drain 


7. Supply Air Distribution: 


A. Duct system; type and general condition 
B. Control box; type and condition 
C. Control box function 


D. Control box/diffuser connection; type and condition 


8. Supply Air Diffusers (see Figure 8-17): 


A. Type and general condition 
B. Characteristics of area served 
C. Number of diffusers this area 
D. Occupied zone air flow pattern; smoke test results 
E. Obstructions to flow pattern 
9. Return Air Grilles (see Figure 8-17): 
A. Type, location, and general condition 
B. Air flow pattern, supply to return; smoke test 


C. Obstructions to flow pattern 


10. Return Air System: 


A. Open plenum; general condition, location of return 
air opening; return air fan/duct 


B. Ducted return; location and general condition 


C. Balancing dampers; type and condition 


11. Miscellaneous Potential Contaminant Sources: 


12. General Comments and Notes: 


8-10 Industrial Ventilation 


REFERENCES 


8.1 The Trane Company: Trane Air Conditioning Manual, 
(February, 1961). 


8.2 International Conference of Building Officials: Light, 
Ventilation and Sanitation, Section 605, Uniform 
Building Code, (1988). 


8.3 American Society of Heating, Refrigeration and Air 
Conditioning Engineers: Ventilation for Acceptable 
Indoor Air Quality, ASHRAE Standard 62-1989. 
ASHRAE, Atlanta, GA (1989). 


8.4 American Society of Heating, Refrigeration and Air 
Conditioning Engineers, Thermal Environmental 
Conditions for Human Occupancy, ASHRAE Stand- 
ard 55-1990. ASHRAE, Atlanta, GA (1992). 


8.5 American Society of Heating, Refrigeration and Air 
Conditioning Engineers: ASHRAE Handbook, 


HVAC Systems and Equipment. ASHRAE, Atltanta, 
GA (1992). 


8.6 Rajhans, G.S.: Findings of the Ontario Inter-Ministe- 
rial Committee on Indoor Air Quality. In: Proceedings 
of the ASHRAE/SOEN Conference, IAQ '89, pp. 195- 
223. ASHRAE, Atlanta, GA (1990). 


8.7 American Society of Heating, Refrigeration and Air 
Conditioning Engineers: ASHRAE Handbook, Fun- 
damentals volume. ASHRAE, Atlanta, GA (1989). 


8.8 Bauer, E.J.; et al.: Use of Particle Counts for Filter 
Evaluation. ASHRAE Journal . ASHRAE, Atlanta, 
GA (October 1973). 


8.9 Duffy, G: Filter Upgrades. Engineered Systems 
(July/August 1993). 


8.10 Ottney, T.C.: Particle Management for HVAC Sys- 
tems. ASHRAE Journal, p.23. ASHRAE, Atlanta, GA 
(July 1993). 


Ventilation Aspects of Indoor Air Quality 8-11 


Reheat coil 


‘X 
& Supply air 


diffusers 
zone #1 


Humidifier— 
Cooling a *e 


| | Supply air 
Seer fan | 


x x (emer (emer: Semeunere —? 
\ 
. Filters ~Condensate pan 


3| Reheat coil -- ( Supply air 


a oa diffusers 


zone #2 


Zone ~ 
thermostat 


Exhaust air 


ee a en en Return air 
eae 


Return air 
fan 


ra 


# = Dampers 


NOTE: See text regarding outside air 
requirements and distribution. 


AMERICAN CONFERENCE 
: OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


SINGLE DUCT CONSTANT 
VOLUME WITH REHEAT 


8-12 industrial Ventilation 


Be Bypass to return air} 


Bypass box — ! 
ne Supply air -7 
diffusers ve 
pda zone ff 
Humidifier 
Cooling oo * 
an \ 


Preheat coil Se 


Outside air | ‘a an thermostat 


Supply air 


—ee-? Bypass to return air 
fan i ; 


\ \ | [Sere CT Ey ET Ne? nikal 
\ Filters \ Condensate pan 7) 


Bypass box — : : 
Supply air 
diffusers 
zone #2 


Zone -~ 
thermostat 


Exhaust air 


othe 


4 ) Return air 
et a yp 


Return air 
fan 


| NOTE: See text regarding outside air 
requirements and distribution. 


AMERICAN CONFERENCE SINGLE. DECT oo 
Fp OOF GOVERNMENTAL | VOLUME WITH BYPAS 
eee! RIAL HYGIENISTS S | “7 Td 


Ventilation Aspects of Indoor Air Quality 8-13 


Reheat coil 


VAV box — 
a % \ Supply air ~ 
\ \ / 


4 diffusers 
Humidifier: . See Veet ae 


Cooling coil~ 


Preheat coil mS 


Zone 
utside air IS f | Tl thermostat 
Co Fone { 


--—- Reheat coil 


“ Condensate pan 


VAV box .. 
Supply air 
diffusers 
zone //1 


Zone~~ 
thermostat 


Return air 


Return air 
VAV fan 


= Dampers 


NOTE: See text regarding outside air 
requirements and distribution. 


AMERICAN CONFERENCE : SINGLE DUCT VARIABLE 
OF GOVERNMENTAL | AIR VOLUME WITH REHEAT 


8-14 Industrial Ventilation 


Induced air —~\ 


VAV induction ope, 
4 Supply air > 
‘ - diffusers 
ale / 
Humidifier [) es 
Cooling coil *\ ‘N 


Zone #1 
Preheat coil e 


| Outside air | y 7 re thermostat 
: Ci 9 


- Induced air 


Condensate pan 


VAV induction box— 
Supply air 
diffusers 
Zone #2 
Zone 
thermostat 


i 


Exhaust air 


Return air 
VAV fan 


NOTE: See text regarding outside air 
requirements and distribution. 


AMERICAN CONFERENCE SINGLE DUCT VARIABLE 
po OOF GOVERNMENTAL AIR VOLUME WITH INDUCTION } 
| INDUSTRIAL HYGIENISTS [ise ——g—9g- rome : 


Ventilation Aspects of Indoor Air Quality 8-15 


Return air — 


Fan powered —\ 
VAV. box sp Supply aire *& 
3 diffusers 
zone #1 
Humidifier \ 


\ 


Cooling coil a 


Preheat coil a4 Nhe 
| 


H Outside air 
: (aan Ss a f 


thermostat 


| 
| eae Return air 
| Su © air 

v VAV fan 


- Filters me 


-Condensate pan 


Fan powered ge ; 
VAV box Supply air 
Son oh diffusers 
pol Zone {2 
Zone ~~ 
thermostat 


Exhaust air 


a) ms 


Return air 
VAV fan 


F NOTE: See text regarding outside air 
: requirements and distribution. 


leet Contetece | SNGlE Duc? Waren aie | 
| AMERICAN CONFERENCE De ees ae er 
a Gn Tk VOLUME WITH FAN POWERED | 


Boe tetera Ae Loa DEVICES AND near 
INDUSTRIAL HYGIENISTS De 4_94 [Fe 


8-16 Industrial Ventilation 


-Cooling coil 

Condensate pan ra 
= ual ducl ; 
( mixing box 


a 


; ; Humidifier f 
Cooling coil ~\ “*< 7 
Preheat coil - \ | i: 
» 


Reheat coil x wey 


\ | 

AW, fo] (D0. Suet air 
T 

S i 7 ? 5 es diffusers 

HOutside air Zone zone #1 

i Oe | 


a Bival duct 
mixing box 


Mites 
Condensate pan -- 


Supply air 
diffusers 
thermostat zone }2 


Exhaust air 


Return. air 


Return air 
fan 


gs = Dampers 


E NOTE: See texi regarding outside air 
: requirements and distribution. 


AMERICAN CONFERENCE | DUAL DUCT CONSTANT 
OF GOVERNMENTAL — | AIR VOLUME 


| INDUSTRIAL HYGIENISTS car 


Ventilation Aspects of Indoor Air Quality 8-17 


Face & by-pass dampers = 
zone control dampers \ 


Heating coil — \ 
Humidifier - . \ —— 
eaeeees te thermostat 
ats ee | 


Preheat coil - 


| Cy i | \/ | Supply air 
roe et ae diffusers 
eo J “\ | Sone a 


a ED 5 


f Outside air 
eg gS 


oe ; i Supply air 
ooling coi we F diffusers 
zone #2 
Condensate pan ’ ; 


6 Zone 
thermostat 


Exhaust air 


ee ee en 


Return air 
fan 


NOTE: See text regarding outside air 
d requirements and distribution. 


SINGLE DUCT CONSTANT 
} AIR VOLUME MULTI-ZONE | 


RENCE 
OF GOVERNMENTAL 
j INDUSTRIAL HYGIENISTS 


| AMERICAN CONFE 


8-18 Industrial Ventilation 


Room supply air 


Wd 


Outside Supply fan 
wall 


Condensate 


pan Cooling coil 


Heating coil 


Filter 


Recirculated air 


WOES SS ouinahaaa) Ce 


Outside Air —> 


(Optional) 
Y 


— Return 
air 


— VE ea 


gs = Dampers 


| AMERICAN CONFERENCE AIR. COIL UNIT 
| OF GOVERNMENTAL | 
| INDUSTRIAL HYGIENISTS 


La ee ae 


Ventilation Aspects of Indoor Air Quality 8-19 


Heat pump zone #1 
(heat or cool) 


Filter 
Supply air 
diffusers 

zone #1 


Humidifier 


Cooling coil 


Preheat coil 


Zone 
thermostat 


Outside air 


a9 


Supply air to 
plenum is ducted 


Puppy air directly to or in 
an near vicinity of 
heat pump 
Filters Condensate pan Supply air 
he diffusers 
zone #2 


AG 


Filter 


Heat pump zone #2 
(heat or cool) 


Zone 
thermostat 


Exhaust air 


os 


Return air 


Return air 
fan 


NOTE: See text regarding outside air 
requirements and distribution. 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


LONE HEAT 
PUMP SYSTEM 


Bm 4-97 pm | 


8-20 Industrial Ventilation 


Ve 


3 
@ 


i 
oO 
te 


EFFICIENCY, 


® 90-35% FICIENCY | 
80-85% EFFICIENCY 
© 30-45% EFFICIENCY 

© 30-45% EFFICIENCY | 


© 25-30% EFFICIENCY 
® 70-80% ARRESTANCE 


PARTICLE SIZE, MICROMETERS 


Approximate Efficiency Versus Particle Size 
for Typical Air Filters (See notes 1 & 2) 


. Compiled and averaged from manufacturer data 
Efficiency and arrestance per ASHRAE Standard 
52 -— 76 Test Methods. 

2. Caution: Curves are approximations only for general guidance. 
Values from them must not be used to specify air filters, 
since a generally recognized test standard does not exist. 


; From: ASHRAE Equipment 


FILTER BFFICIENCY 


| VS PARTICLE SIZE 


AMERICAN CONFERENCE 
: OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


Ventilation Aspects of Indoor Air Quality 8-21 


if 
: 
| ie 


PARALLEL BLADE 


: DAMPERS: 
PARALLEL AND OPPOSED BLADE 


AMERICAN CONFERENCE | 
OF GOVERNMENTAL 
1} INDUSTRIAL HYGIENISTS 


Industrial Ventilation 


8-22 


Disposable Media : 


Filter 


LL 
art 


tr 
a 
Ba 
=e 
as 
S 
ee 
— 
= 
o 
= 
— 


ficiency 
ct 
th 


Ef 
Disposable 
ell Fitter 


vas 


f 
Me 


High 
Filter 
NT 


ERENC 
AL 


ti 


4 


GOVERNME. 


NF 


or 
Washable Metal 
CON 


AN 


N 
/ 


Rl 


OF 


uper Inception 


a 
a 


Absolute Filter 
REC 


S 


AME 


Ventilation Aspects of Indoor Air Quality 8-23 


COILS 


CONFERENCE HEATING /COOLING 
OVERNMENTAL 
TRIAL HYGIENISTS 


8-24 Industrial Ventilation 


HOUSING 


DIVERTER 


SIDE SHEET 


BACKPLATE 


BEARING 
SUPPORT 


INLET COLLAR 


Reprinted from AMCA Publication 201-90, FANS AND SYSTEMS, by 
permission of the Air Movement and Control Association, Inc.©) 


TERMINOLOGY FOR | 
CENTRIFUGAL FAN 
COMPONENTS id 


| AMERICAN CONFERENCE | 
| OF GOVERNMENTAL | 
| INDUSTRIAL HYGIENISTS [igg-—p—55——" rome 


Ventilation Aspects of Indoor Air Quality 8-25 


V 


} 


) WET AIR IN gy | 


{ 


CONDENSATE PAN 


REFRIGERATION DEHUMIDIFICATION 


ELIMINATORS- 


ATOMZING HUMIDIFICR WITRE OPTIONAL 
FILTER ELIMINATOR 


PNEUMATIC ATOMIZING HUMIDIFIER 


AMERICAN CONFERENCE | HVAC 
OF GOVERNMENTAL J HUMIDIFIER \ DEHUMIDIFIER 


8-26 Industrial Ventilation 


ROOF 


FILTER SECTION 


HUMIDIFICATION SECTION 
FAN SECTION 


4 es 2 
Se PREHEAT COILS REHEAT COIL 


MIXING SECTION DIFFUSER PLATE 
DAMPER SECTION FLOOR 


HVAC SELF CONTAINED 


PENTHOUSE ROOF 


FAN SECTION 


PRE FILTER 
FINAL FILTER 
HEATING COIL 
COOLING COIL 


LOX 


DAMPER SECTION > 
7 / 
Perey j 7 ROOF “ 


RETURN AIR 


SUPPLY AIR 
SINGLE DUCT 


HVAC CENTRAL EQUIPMENT _ ROOM — 


I 
AMERK C AN CONFERENCE SELF CONTAINED 


f OF GOVERNMENTAL | — anpD EQUIPMENT ROOM 
| INDUSTRIAL HYGIENISTS [aE—g—9gF 


Ventilation Aspects of Indoor Air Quality 8-27 


SIDE WALL ADJUSTABLE 


AIR. SUPPLY DIFFUSERS 


XETURN AIR GRILLES 


Alf SUPPLY DIFFUSE NS <& 
REPURN: Ale GALES 
FIGURE 20 


AMERICAN CONFERENCE | 
OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


8-28 Industrial Ventilation 


IDE WALL “SUPELY 


| AMERICAN CONFERENCE | TYPICAL PARTITIONED 
POF GOVERNMENTAL, | OFFICE AIR PATTERN 
J INDUSTRIAL HYGIENISTS [3-99-77] 


Chapter 9 


TESTING OF VENTILATION SYSTEMS 


9.1 INTRODUCTION .................... 9-2 9.4.2 Hood Static Pressure... 2... 2. 
9:1 + Initial Lest. 9: xno bay Pek Ed 9-2 9.4.3 Hood Static Pressure Interpretation ...... 
9.1.2 PeriodicTest .. 2... 0.062 eeu eeeae 9-2 9.44 Velocity Pressure .............0.. 
9.2. MEASUREMENTS OF VOLUMETRIC 9.5  PITOTTRAVERSE METHOD ............ 
FLOWRATE .. 0... 602 ee ee ees 9-2 96 CORRECTIONS FOR NON-STANDARD 
9.2.1. Air Velocity Instruments ............. 9-3 CONDITIONS ..............-02.205- 
9.3. CALIBRATION OF AIR MEASURING 9.6.1 Example Traverse Calculations ....... 
INSTRUMENTS .. 2... 6.20 ee ee eee 9-6 9.7 CHECK-OUT PROCEDURES............ 
9.3.1 Design of a Calibrating Wind Tunnel .... . 9-6 9.7.1 Difficulties Encountered in Field 
9.3.2 Use of Calibrating Wind Tunnel ........ 9-7 Measurement ..........-..--- 
9.4 PRESSURE MEASUREMENT ............ 9-11 REFERENCES.) ooo: 54: bod a tek oe Beeld tenon oe eae 
9.4.1 Static Pressure... 2... ...0.0...22.0. 9-11 
Figure 9-1 Rotating Vane Anemometer ........... 9-3 Figure 9-10A 10-Point Pitot Traverse in a Circular Duct . . . 
Figure 9-2 Swinging Vane Anemometer... ........ 9-4 Figure 9-10B 6-Point Pitot Traverse in a Circular Duct... . 
Figure 9-3 Anemometer Applications ............ 9-5 Figure 9-11 Velocity Pressure Distributions... ...... 
Figure 9-4 Thermal Anemometer... ..........-. 9-6 Figure 9-12 Pitot Traverse Points in a Rectangular Duct’ . . 
Figure 9-5 Calibration Wind Tunnel .. 2... 2.020. . 9-8 Figure 9-13 U-Tube Manometer ............0.0. 
Figure 9-6 Calibration: ois jo-0, t8 Sets eke. enh as 9-9 Figure 9-14 Inclined Manometer............0.. 
Figure 9-7 Static Tap Connections .............. 9-11 Figure 9-15 AneroidGauge ..............05. 
Figure 9-8 Pitot Tube Measurement ............. 9-12 Figure 9-16 Sample System ...........-..0-. 
Figure 9-9 Standard Pitot Tube... 2... 9-14 Figure 9-17 SurveyForm.................-. 


9-2 Industrial Ventilation 


9.1 INTRODUCTION 


Every ventilation system should be tested at the time of 
initial installation to verify the volumetric flow rate(s) and to 
obtain other information which can be compared with the 
original design data. Testing is necessary to verify the setting 
of blast gates, fire dampers, and other air flow control devices 
which may be a part of the system. Initial testing will provide 
a baseline for periodic maintenance checks and isolation of 
system failures should a malfunction occur. Many govern- 
mental codes require initial and periodic testing of exhaust 
systems for certain types of processes. Exhaust system test 
data are also useful as a basis for design of future installations 
where satisfactory air contaminant control is currently being 
achieved. 


The tests described in this text pertain to ventilation sys- 
tems only. Environmental! tests should be conducted prior to 
and after installation to verify system performance. In these 
cases, the services of a qualified industrial hygienist may be 
required. 


9.1.1 Initial Test: The Pitot tube and manometer are the 
standard for initial field testing of equipment when used as 
described. However, other instruments may be used. As noted 
later, all instruments must be calibrated. Identify on the test 
sheet the instruments and procedures used. A sample survey 
form is located at the end of the chapter (Figure 9-17). The 
following steps outline the recommended procedure and the 
minimum data necessary fora thorough initial ventilation test. 


1. Review the system specifications and drawings to de- 
termine the relative location and sizes of ducts, fittings, 
and associated system components. Where possible, 
pertinent prints should be carried to the test site. 


i) 


. Inspect the system to determine that its installation is 
in accordance with the specifications and drawings. 
Check such items as fan rotation, belt slippage, and 
damper settings. 


3. Include a drawing of the system as installed. Select and 
identify test locations. 


4. Measure the volumetric flow rate, fan static pressure, 
fan speed, motor speed, motor amperes, and the tem- 
perature of the air in the system. Also determine pres- 
sure drops across all components such as coils, fittings, 
and air cleaning equipment. 


Fan speed may be measured directly at the end of the 
fan shaft using a revolution counter and a watch. A 
tachometer or stroboscopic measuring device also may 
be used. 


The operating amperage is obtained with an ammeter. 
The readings taken on each lead on the three-phase 
current should be averaged and compared to the rating 
on the motor name plate at the operating voltage to 
determine if the motor is operating within its rated 


range. 


For some tests, moisture content of the air in the system 
and/or ambient barometric pressure should be obtained 
also. See example calculations in Section 9.6. 


5. Record the test data and design specifications on the 
data sheet. Calculate test results following the format 
on the data sheets. 


6. Compare the test data with design specifications. De- 
termine if alterations or adjustments of the system are 
necessary to meet specifications, codes, or standards. 


7. If alterations or adjustments are made, retest the sys- 
tem, and record the final test data. On the drawing, note 
any physical changes that were made in the system. 


9.1.2 Periodic Test: The performance of a system should 
be checked periodically. If there have been no alterations to 
the system, this can be done by static pressure measurements 
and close visual inspection. Measurements also can be made 
continuously by means of an operating console or other 
remote readout system. 


The following is the recommended procedure with sugges- 
tions as to types of measurements needed to perform the 
periodic tests: 


1. Refer to the initial data sheet for test locations. 


2. Inspect the system for physical damage (broken, cor- 
roded, collapsed duct, etc.) and proper operation of com- 
ponents (fan, damper, air cleaner, controls, burner, etc.). 


3. Measure static pressure at the same locations used in 
the initial test. 


4. Compare measured static pressures with initial test. 
From these comparisons, determine if the system is 
performing at initial levels. 


5. Make and record any correction required. 


6. Recheck the system to verify performance if correc- 
tions have been made. 


Whenever alterations have been made to the system, a new 
initial test is necessary following the procedures outlined 
under Section 9.1.1, "Initial Test." 


9.2 MEASUREMENTS OF VOLUMETRIC FLOW RATE 


The most important measurement in testing of systems is 
the measurement of the volumetric flow rate in cfm. This 
should be done before balancing the system is attempted. The 
commonly used instruments are of the velocity measuring 
type rather than quantity meters. Therefore, it is necessary to 
obtain not only the average air velocity through an opening 
or duct, but also the net cross-sectional area at the point of 
measurement. The volumetric flow rate then can be deter- 
mined from the equation: 


Q=VA [9.1] 


TABLE 9-1. Characteristics of Flow Instruments 


Testing of Ventilation Systems 9-3 


General 
Hole Size Range, Dust, Fume Calibration Usefulness 
Instrument Range, fpm (for ducts) Temp* Difficulty Requirements Ruggedness and Comments 
PITOT TUBES with inclined manometer 
Standard 600 - up 3/8" Wide Some None Good Good except at 
low velocities 
Small Size 600 - up 3/16" Wide Yes Once Good Good except at 
low velocities 
Double 500 — up 3/4" Wide Small Once Good Dirty air stream 
SWINGING VANE ANEMOMETERS 
25 — 10,000 1/2 - 1" Medium Some Frequent Fair Good 
ROTATING VANE ANEMOMETERS 
Mechanical 30 - 10,000 Not for duct use Narrow Yes Frequent Poor Special; limited 
use 
Electronic 25 — 200 Not for duct use Narrow Yes Frequent Poor Special; can 
25 - 500 record; direct 
25 — 2000 reading 
25 - 5000 


*Temperature range: Narrow, 20-150 F; Medium, 20-300 F; Wide, 0-800 F 


where: 
Q= volumetric flow rate, (cfm) 
V = average linear velocity, feet per minute (fpm) 


A= cross-sectional area of duct or hood at the meas- 
urement location, ft® 


9.2.1 Air Velocity Instruments: The volumetric flow rate 
of an exhaust system can be determined by the use of various 
types of field instruments which measure air velocity directly. 
Typically, these instruments are used at exhaust and discharge 
openings or, depending on size and accessibility, inside a duct. 
The field technique is based on measuring air velocities at a 
number of points in a plane and averaging the results. The 
average velocity is used in Equation 9.1 to determine the 
volumetric flow rate. Due to the difficulty of measuring the 
area of an irregularly shaped cross-section and the rapid 
change in velocity as air approaches an exhaust opening, meas- 
urements obtained should be considered an approximation of the 
true air flow. All instruments should be handled and used in strict 
compliance with the recommendations and directions of the 
manufacturers. Table 9-1 lists some characteristics of typical air 
velocity instruments designed for field use. 


Rotating Vane Anemometer (Figure 9-1): This instrument 
is accurate and can be used to determine air flow through large 
supply and exhaust openings. Where possible, the cross-sec- 
tional area of the instrument should not exceed 5.0% of the 
cross-sectional area of the duct or hood opening. The standard 
instrument consists of a propeller or revolving vane connected 
through a gear train to a set of recording dials that read the 
linear feet of air passing in a measured length of time. It is 
made in various sizes; 3", 4", and 6" are the most common. It 


gives average flow for the time of the test (usually one 
minute). The instrument requires frequent calibration and the 
use of a calibration card or curve to determine actual velocity. 
The instrument may be used for either pressure or suction 
measurements using the correction coefficients listed by the 
manufacturer. The standard instrument has a useful range of 
200-3000 fpm; specially built models will read lower velocities. 


Direct-recording and direct-reading rotating vane ane- 
mometers are available. These instruments record and meter 


FIGURE 9-1. Rotating vane anemometer 


9.4 Industrial Ventilation 


electrical pulses developed by a capacitance or inductive 
transducer. The impulses are fed to the indicator unit where 
they are integrated to operate a conventional meter dial. 
Readings as low as 25 fpm can be measured and recorded. 


The standard 4" rotating vane anemometer is unsuited for 
measurement in ducts less than 20" in diameter as it has too 
large a finite area and its equivalent cross-sectional area is 
difficult to compute. The conventional meter is not a direct- 
reading velocity meter and must be timed. It is fragile and care 
must be used in dusty or corrosive atmospheres. Newer units 
of 1" diameter which can be used in ducts as small as 5" in 
diameter are available. 


Swinging Vane Anemometer (Figure 9-2): This instrument 
is extensively used in field measurements because of its 
portability, wide-scale range, and instantaneous reading fea- 
tures. Where accurate readings are desired, the correction 
coefficients in Table 9-2 should be applied. The instrument 
has wide application and, by a variety of fittings, can be used 
to check static pressures and a wide range of linear velocities. 
The minimum velocity is 50 fpm unless specially adapted for a 
lower range. The instrument is fairly rugged and accuracy is 
suitable for most field checks. Uses of the swinging vane ane- 
mometer and its various fittings are illustrated in Figure 9-3. 


Before using, check the meter for zero setting by holding 
it horizontal and covering both ports so that no air can flow 
through. If the pointer does not come to rest at zero, an 
adjustment must be made to correct the starting point. The 
meter should be used in an upright position and, when using 
fittings, it must be held out of the air stream so that the air 
flows freely into the opening. The length and inside diameter 
of the connecting tubing will affect the calibration of the 
meter. When replacement is required, use only connecting 
tubing of the same length and inside diameter as that origi- 
nally supplied with the meter. 


_ 


FIGURE 9-2. Swinging vane anemometer 


TABLE 9-2. Correction Factors for the Swinging Vane and Thermal 
Anemometers 


Correction Factor C, 


Grille Openings (percent) 
Pressure 

More than 4 in wide and up to 600 in? area, 

free opening 70% or more of gross area, no 

directional vanes. Use free-open area. 93 
Suction 

Square punched grille (use free-open area) 88 

Bar grille (use gross area) 78 

Strip grille (use gross area) 73 


Free open, no grille No correction 


Where temperatures of an air stream vary more than 30 F 
from the standard temperature of 70 F and/or if the altitude is 
greater than 1,000 feet, it is advisable to make a correction for 
temperature and pressure. Corrections for change in density 
from variations in altitude and temperature can be made by 
using the actual gas density (p) shown in Equation 9.9 in the 
following equation: 


ee alls [9.2] 
p 
where: 


V, = corrected velocity, fpm 


< 
" 


velocity reading of instrument, fpm 


p = actual gas density, Ibm/ft? 


Use at Supply Openings: On large (at least 3 ft?) supply 
openings where the instrument itself will not block the open- 
ing seriously and where the velocities are low, the instrument 
itself may be held in the air stream with the air impinging 
directly in the left port. When the opening is smaller than 3 
ft? and/or where the velocities are above the "No Jet" scale, 
appropriate fittings must be used. 


Because the velocity and static gradient in front of an 
exhaust opening is steep, the finned opening of the fitting must 
be held flush with the exhaust opening. If the opening is 
covered by a grille, hold the fin directly against the grille and 
use the correction coefficients listed in Table 9-2 when com- 
puting exhaust volumes. 


Q=C,VA [9.3] 


where: 
C, = correction coefficient in percent of scale read- 
ing. 

While it can be used to measure air velocities, static pres- 
sure, and total pressure in ducts, it has several disadvantages. 
Used in place of a Pitot tube for velocity or total pressure 
measurements, it necessitates a much larger hole in the duct, 
often difficult and impractical to provide. When the velocities 


Testing of Ventilation Systems 9-5 


DIFFUSER 


BOXBOARD CONE 
" (COMMERCIALLY AVAILABLE) 


AREA SIZED FOR 
MAXIMUM VELOCITY 


OF 400 FPM 


| ANEMOMETER 
| APPLICATIONS 


| AMERICAN CONFERENCE 
| OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


9-6 Industrial Ventilation 


are high, there may be no appreciable errors at the high end 
of the scale and the instrument tends to read low on the 
discharge side of the fan and high on the inlet side. 


The presence of dust, moisture, or corrosive materia] in the 
atmosphere presents a problem since the air passes through 
the instrument. In those instruments calibrated for use with a 
filter (the filter must always be used), the filter itself is a source 
of error because as the filter becomes plugged, its resistance 
increases and thus alters the amount of air passed to the 
swinging vane. The instrument requires periodic calibration 
and adjustment. 


Thermal Anemometer (Figure 9-4): This type of instrument 
employs the principle that the amount of heat removed by an 
air stream passing a heated object is related to the velocity of 
the air stream. Since heat transfer to the air is a function of the 
number of molecules of air moving by a fixed monitoring 
point, the sensing element can be calibrated as a mass flow 
meter as well as a velocity recorder. Commercial instruments 
use a probe which consists of two integral sensors: a velocity 
sensor and a temperature sensor. The velocity sensor operates 
at a constant temperature —- typically about 75 F above 
ambient conditions. Heating energy is supplied and controlled 
electrically by a battery-powered amplifier in the electronics 
circuit. The electrical current required to maintain the probe 
temperature in conjunction with the temperature sensor will 
provide an electrical signal which is proportional to the air 
velocity and is displayed on either a digital or analog meter. 
Additional features often include time integration of fluctuat- 
ing readings and air temperature at the probe. Displays are 
available in either English or S.1. units. 


The velocity sensor should be used with care in normal field 
use and is insensitive to mild particulate contamination. The 
probe can be used directly to measure air velocity in open 
spaces at air exhaust and supply air openings. Attachments 


FIGURE 9-4. Thermal Anemometer 


are available to measure velocity pressure. Due to the small 
diameter of the probe, measurements can be made directly 
inside of ducts using the measurement techniques described 
later for Pitot traverses. When used at supply or exhaust 
openings covered by grilles, the correction coefficient listed 
in Table 9.2 should be used. 


Battery charging and maintenance is extremely important 
and the battery voltage must be checked prior to instrument 
use. The correction coefficients for this instrument are the 
same as a Swinging vane anemometer (see Table 9.2). Instru- 
ments of this type require both initial and periodic calibration. 


Smoke Tubes: Low velocity measurements may be made 
by timing the travel of smoke clouds through a known dis- 
tance. Smoke trail observations are limited to velocities less 
than 150 fpm since high air velocities diffuse the smoke too 
rapidly. Commercially available, smoke tubes and candles are 
useful in the observation of flow patterns surrounding exhaust 
or supply openings. They also can be used for checking air 
movement and direction in plant space. 


The visible plume is corrosive and should be used with care 
near sensitive processes or food preparation. Smoke candles 
are incendiary and thus cannot be used in flammable atmos- 
pheres. They should not be hand-held. Alternative methods of 
observing air flow patterns include the use of soap bubbles, water 
vapor cooled by dry ice (CO2), and heated vegetable oil. 


Tracer Gas: The principle of dilution sometimes is used to 
determine rate of air flow. A tracer gas is metered continu- 
ously into one or more intake ports (hood or duct openings) 
along with the entering air stream. After thorough mixing and 
system equilibrium has been established, air samples are 
collected at some point downstream — usually at or near the 
effluent point — and the concentration of the tracer gas in the 
exit stream is determined. The rate of air flow is readily 
calculated from the degree of dilution noted in the exit and 
feed gas concentrations (rate of air flow equals rate of feed 
divided by tracer gas concentrations).°°) 


The tracer gas usually is selected on the basis of the 
following: 1) ease of collection and analysis, 2) not present 
naturally in the process being studied, 3) not absorbed chemi- 
cally or physically in the duct system, 4) non-reactive with 
other constituents of the gas stream, and 5) non-toxic, non- 
explosive, and non-odorous. Some frequently used tracer 
gases are sulfur hexafluoride and carbon dioxide. 


9.3 CALIBRATION OF AIR MEASURING INSTRUMENTS 


Direct-reading meters need regular calibration because 
they can be easily impaired by shock (dropping, jarring), dust, 
high temperatures, and corrosive atmospheres. Meters should 
be calibrated regularly and must be calibrated if they will not 
adjust to zero properly or if they have been subjected to rough 
handling and adverse atmospheres. 


9.3.1 Design of a Calibrating Wind Tunnel: A typical 


calibrating wind tunnel for testing air flow meters must have 
the following components: 


1. A satisfactory test section. This is the section where 
the sensing probe or instrument is placed; it must be 
uniform in air flow both across the air stream and in 
line with the air flow. A section with a pronounced 
vena contracta and turbulence will not give satisfac- 
tory results. 


2. A satisfactory means of precisely metering the air flow. 
The meter on this system must be accurate and with 
large enough scale graduations so that the volumetric 
flow rate is indicated within + 1%. For convenience 
and time saving, a fixed single reading meter such as 
a venturi meter or orifice meter is preferable to a 
multi-point traverse type instrument such as a Pitot 
tube. 


3. A means of regulating and effecting air flow through 
the tunnel. For usual calibrations of instruments used 
on heating, ventilating, and industrial exhaust systems, 
test velocities from approximately 50 to 8000 fpm are 
needed. Air flow regulation must be such that there is 
no disturbance in the test section. The regulating de- 
vice must be easily and precisely set to the desired 
velocities. The fan must have sufficient capacity to 
develop the maximum velocity in the test section 
against the static pressure of the entire system. 


To provide a satisfactory uniform flow in the test section, 
a bell-shaped streamline entry is necessary (Figure 9-5). There 
are various designs for this entry. One type is the elliptical 
approach in which curvature is similar to a one-quarter section 
of an ellipse in which the semi-major axis of the ellipse is 
equal to the duct diameter to which the entry is placed and the 
semi-minor axis is two-thirds of the semi-major axis. This 
type of entry can be made on a spinning lathe. 


Actually, any type of smooth curved, bell-shaped entry 
which directs the air into the duct over a 180° angle should be 
satisfactory. A readily available entry is atuba or Sousaphone 
bell. This bell entry should be connected to a 5.5" diameter 
smooth, seamless plastic tube. Ridges, small burrs, or obstruc- 
tions should be filed so a smooth connection between horn 
and tube results. 


For calibrating larger instruments such as the lower veloc- 
ity swinging vane anemometer (Alnor velometer) and the 
rotating vane anemometer, a large rectangular test section of 
transparent plastic at least 2.5 ft? in cross-sectional area can 
be constructed with curved air foil inlets as shown in Figure 
9-6. A fine mesh screen placed deep in the enclosure will assist 
in providing a uniform air flow in the test section. 


A sharp-edged orifice, venturi meter, or a flow nozzle can 
be used as a metering device. Of these, the sharp-edged orifice 
has more resistance to flow but is more easily constructed, 
and it can be designed to be readily interchangeable for several 


Testing of Ventilation Systems 9-7 


orifice sizes. The orifice can be mounted between two flanged 
sections sealed with gaskets as shown in Figure 9-5. Each 
orifice should be calibrated using a standard Pitot tube and 
manometer prior to use. For velocity measurements below 
2,000 fpm, a micromanometer should be used.© 


Table 9-3 lists calculations for three sizes or orifices: 
1.400", 2.625" and 4.900" diameters. When the orifices are 
placed in a 7" diameter duct and made to the precise dimen- 
sions given, no calibration is needed and the tabulated data in 
the Table will give volumetric flow rates within + 5% over 
the range of values shown for standard air density. 


A centrifugal fan with sufficient capacity to exhaust |, 100 
cfm at 10 "wg static pressure is needed for a wind tunnel with 
a5.5" diameter test section using an orifice meter. Radial and 
backwardly inclined blade centrifugal fans are available with 
the required characteristics. The air flow can be changed with 
an adjustable damper at the discharge, a variable speed motor, 
or an adjustable drive on the fan. 


The air flow for a sharp-edged orifice with pipe taps located 
1" on either side of the orifice can be computed from the 
following equation for 2"-14" diameter ducts: 


a =eKp? {° [9.4] 
p 


where: 
Q = volumetric flow rate, cfm 
K = coefficient of air flow 
D = orifice diameter, inches 
h = pressure drop across orifice, "wg 
p= density, Ib/m/ft? 
The coefficient, K, is affected by the Reynolds number — 
a dimensionless value expressing flow conditions in a duct. 


The following equation gives a simplified method of calcu- 
lating Reynolds number for standard air: 


R=84DV [9.5] 


where: 
R= Reynolds number, dimensionless 
V= velocity of air through orifice, fpm 


The coefficient, K, can be selected from Table 9-4.0*) 


9.3.2 Use of Calibrating Wind Tunnel: Air velocity meas- 
uring instruments must be calibrated in the manner in which 
they are to be used in the field. Swinging vane and rotating 
vane anemometers are placed in the appropriate test section 
on a suitable support and the air velocity varied through the 
operating range of interest. Heated thermocouple instruments 
are calibrated in the same manner. Special Pitot tubes and duct 
probes of direct-reading instruments are placed through a 
suitable port in the circular duct section and the air velocity 
varied through the operating range of interest. Heated thermo- 


9-8 Industrial Ventilation 


3 HP MOTOR WITH VARIABLE DRIVE 
500 TO 3670 RPM 
re 


/ . STREAMLINE INLET ALTERNATE DAMPER 


PLASTIC TUBE 


STRAIGHTENERS ~ 


TEST SECTION 1 
FOR HIGH VELOCITY METERS MANOMETER - 6” INCLINE | § 
WITH SMALL TEST PROBES 


15° VERTICAL) ! 
IN TEST AIR STREAM. ; 


CALIBRATION WIND TUNNEL 


PIPE TAPS — 


—— Ee 


| Sea aC 


SHARP EDGE ORIFICE ~ 


1/8 STEEL PLATE GASKET 


ORIFICE DISTAL, 


BRACKET ON ROD 
TRANSPARENT PLASTIC 
TEST SECTION 


FOR LOW VELOCITY METERS WITH 
LARGE AREA IN TEST AIR STREAM. 


AMBRICAN CONFERENCE he sgeat: error 

: [ ID TUNNE 

| oF GOVERNMENTAL | CALIBRATION WIND TUNNEL 
| INDUSTRIAL HYGIENISTS 


Testing of Ventilation Systems 9-9 


15° OR MORE =—=— 12” OR LESS —=| 


HEATED THERMOCOUPLE PROBE IN TEST SECTION 
\ 


Ima 15” OR MORE D>’ 
IF STAND {S 


SCREEN 


| 


7 — BRACKET 
~ TEST SECTION 


LARGE AIR METER IN TEST SECTION 


KEEP TEST SECTION ENTRANCE 
CLEAR OF OBSTRUCTIONS AND 
FREE OF DRAFTS 


AMERICAN CONFERENCE Pe ee 
PQ] / yt 
OF GOVERNMENTAL eee a 


INDUSTRIAL HYGIENISTS k= 


9-10 Industrial Ventilation 


TABLE 9-3. Orifice Flow Rate (scfm) Versus Pressure Differential (in. of water) 


ORIFICE SIZE ORIFICE SIZE ORIFICE SIZE 
1.4" 2.625" 4.90” 2.625" 4.90" 2.625” 4.90” 


57.1 101.4 410.3 185.3 746 
18.7 78.8 102.3 413.6 187.5 755 
22.8 95.3 103.1 416.9 189.7 763 
26.2 109.2 103.9 420.1 191.9 772 
29.3 121.5 104.7 423.4 194.0 781 
32.1 132.6 105.5 426.5 196.2 789 
34.6 142.8 106.3 429.7 : 198.3 797 
37.0 152.3 107.1 432.9 200.4 
39.2 161.2 107.9 436.0 202.4 
41.3 169.6 108.6 439.1 204.4 
43.3 177.6 109.4 442.2 206.5 
45.2 185.2 110.2 445.2 208.5 
47.0 192.6 110.9 448.3 : 210.4 
48.8 199.6 111.7 451.3 212.4 
50.5 206.5 112.4 454.3 214.3 
52.1 213.0 113.2 457.2 216.3 
53.7 219.4 113.9 460.2 218.2 
55.3 225.6 114.6 463.1 220.0 
56.8 231.6 115.4 466.0 : 221.9 
58.3 237.5 116.1 468.9 223.8 
59.7 243.2 116.8 471.8 225.6 
61.1 248.8 117.5 474.7 227.4 
62.4 254.3 118.2 477.5 229.2 
63.8 259.6 : 118.9 480.3 231.0 
65.1 264.9 : 119.6 483.1 232.8 
66.4 270.0 120.3 485.9 : : 234.6 
67.6 275.0 121.0 488.7 : 236.3 
68.9 280.0 121.7 491.5 238.1 
70.1 284.8 , 122.4 494.2 239.8 
71.3 289.6 : 123.1 496.9 241.5 
72.4 294.3 : 123.8 499.7 . ; 243.2 
73.6 298.9 é 124.4 502.4 ; 244.9 
74.7 303.4 : 125.1 505.0 : 246.5 
75.8 307.9 125.8 507.7 248.2 
76.9 312.3 126.4 510.4 249.9 
78.0 316.7 127.1 513.0 251.5 
79.1 320.9 127.8 515.6 253.1 
80.2 325.2 128.4 518.2 254.7 
81.2 329.3 129.1 520.8 256.4 
82.2 333.5 129.7 523.4 257.9 
83.2 337.5 132.9 536.2 259.5 
84.2 341.6 136.0 548.6 261.1 
85.2 345.5 139.0 560.8 262.7 
86.2 349.4 142.0 572.6 264.2 
87.2 353.3 144.9 584.3 265.8 
88.1 357.2 147.8 595.7 267.3 
89.1 361.0 150.6 606.9 268.8 
90.0 364.7 153.3 617.9 270.4 
91.0 368.4 156.0 628.6 271.9 
91.9 372.1 158.7 639.2 273.4 
92.8 375.7 ; 161.3 649.6 : 274.9 
93.7 379.3 163.8 659.9 276.4 
94.6 382.9 166.4 670.0 277.8 
95.5 386.4 168.8 679.9 279.3 
96.3 390.0 171.3 689.7 280.8 
97.2 393.4 173.7 699.3 282.2 
98.1 396.9 170.1 708.8 283.6 
98.9 400.3 178.4 718.2 285.1 
99.8 403.7 é : 180.7 727.5 : 286.5 
100.6 407.0 183.0 736.6 287.9 


Testing of Ventilation Systems 9-11 


RUBBER HOSE. 


DUCT WALL — DRILL ALL HOLES 1/16” D. OR LESS. MAINTAIN 
INNER SURFACE OF DUCT SMOOTH AND FLUSH. 


ONE HOLE RUBBER STOPPER, CONNECTING TUBE AND 


PERMANENT TAP -—1/8" TEE WITH 1/8" PIPE PLUG, T-HANDLE AND 
CLEANING WIRE. EQUIP WITH 1/8” PLUG WHEN NOT IN USE. USE 
1/8" COUPLING IF USED FOR PERMANENT CONNECTION TO U-TUBE. 


1/8" PETCOCK AND COUPLING WITH 3/16” COPPER TUBE 
SOLDERED TO PETCOCK. USE CAREFULLY WHEN HIGH 
PRESSURES ARE MEASURED TO PREVENT GAUGE LIQUID 
SURGE ON OPENING COCK. 


FIGURE 9-7, Static tap connections 


couple instruments are calibrated in the same manner. Special 
Pitot tubes and duct probes of direct-reading instruments are 
placed through a suitable port in the circular duct section and 
calibrated throughout the operating range (Figure 9-6).0” 


9.4 PRESSURE MEASUREMENT 


At any point in an exhaust system, three air pressures exist 
which can be compared to the atmospheric pressure immedi- 
ately surrounding the system. Typically, these pressures are 
measured in inches water gauge ("wg) and are related to each 
other as follows: 


TP =SP+VP [9.6] 


where: 
TP = total pressure, "wg 
SP = static pressure, "wg 
VP = velocity pressure, "wg 


Static pressure is that pressure which tends to burst or 
collapse a duct and is positive when the pressure is above 
atmospheric and negative when below atmospheric. Velocity 
pressure is the pressure resulting from the movement of air 
and is always positive. Total pressure is the algebraic sum of 
the static pressure and velocity pressure and can be either 
positive or negative (see Figure 9-8). 


9.4.1 Static Pressure is measured by a pressure measur- 
ing device, usually a simple U-tube manometer filled with oil, 
water, or other appropriate liquid and graduated in inches 
water gauge or similar reading pressure gauge. A vertical 
manometer is suitable for most static pressure measurements. 
The use of an inclined manometer will give increased accu- 
racy and permits reading of lower values. For field measure- 
ment, one leg of the manometer is open to the atmosphere and 


the other leg is connected with tubing held flush and tight 
against a small opening in the side of the pipe. Additional 
information concerning manometers and their construction 
can be found in References 9.1 and 9.2. 


The location of the static pressure opening is usually not 
too important in obtaining a correct measurement except that 
one should avoid pressure measurement at the heel of an 
elbow or other location where static pressure may be incorrect 
because the direction of the velocity component is not parallel 
with the duct wall. It is usually advisable to drill 2-4 pressure 
holes at uniform distances around the duct in order to obtain 
an average and to detect any discrepancy in value. 


The static pressure opening should be flush with the inner 
surface of the pipe wall and there should be no burrs or 
projections on the inner surface. The hole should be drilled, 
not punched. A 1/16"—1/8" hole is usually satisfactory since 
the size is not too important except for some types of instru- 
ments where air actually flows through the device (see Figure 
9-7). The recommendations of the manufacturer concerning 
the size of the static pressure opening should be followed. A 
second method less likely to involve error is to use the static 
pressure element of a Pitot tube as shown in Figure 9-8. In 
use, the instrument must be pointed upstream and parallel to 
the duct for accurate measurement. 


9.4.2 Hood Static Pressure: The hood static pressure 
method of estimating air flow into an exhaust hood or duct is 
based on the principle of the orifice; i-e., the inlet opening 
simulating an orifice. This method is quick, simple, and 
practical. It is a fairly accurate estimation of the volumetric 
air flow in branch exhaust ducts ifthe static pressure or suction 
measurement can be made at a point one to three duct diame- 
ters of straight duct downstream from the throat of the exhaust 


9-12 Industrial Ventilation 


TOTAL PRESSURE = STATIC PRESSURE + VELOCITY PRESSURE 


TOTAL PRESSURE | 
BELOW ATMOSPHERE j ay i 


STATIC PRESSURE 
BELOW ATMOSPHERE 


VELOCITY PRESSURE 
ABOVE ATMOSPHERE 


AMIBIIC AN SCONPER ENGI itis ot ise. pry erie a cee 
| JOP COvBRNMENTAL. |, oe Pee eee ene 


| INDUSTRIAL 


HYGIENISTS ] 


Testing of Ventilation Systems 9-13 


TABLE 9—4. Values of K in Equation 9.10 for Different Orifice Diameters to Duct-Diameter Ratios (d/D) and Different Reynolds Numbers* 
Reynolds Number in Thousands 


d/D 25 50 100 230 500 1000 10,000 
0.100 0.605 0.601 0,598 0.597 0.596 0,595 0.595 
0.200 0.607 0.603 0.600 0.599 0.598 0.597 0.597 
0.300 0.611 0.606 0.603 0.603 0.601 0.600 0.600 
0.400 0.621 0.615 0.614 0.610 0.609 0.608 0.608 
0.450 0.631 0.624 0.619 0.617 0.615 0.615 0.615 
0.500 0.644 0.634 0.628 0.626 0.624 0.623 0.623 
0.550 0.663 0.649 0.641 0.637 0.635 0.634 0.634 
0.600 0.686 0.668 0.658 0.653 0.650 0.649 0.649 
0.650 0.717 0.695 0.680 0.674 0.670 0.668 0.667 
0.700 0.755 0.723 0.707 0.699 0.694 0.692 0.691 
0.750 0.826 0.773 0.747 0.734 0.726 0.723 0.721 

*For duct diameters of 2" to 14” inclusive. 

inlet and if an accurate analysis of the hood entry loss can be Q = 4005AC, JSP, [9.8] 


made. 


This technique involves the measuring of hood static pres- 
sure by means of a U-tube manometer at one or more holes 
(preferably four, spaced 90° apart), one duct diameter down- 
stream from the throat for all hoods having tapers, and three 
duct diameters from the throat for flanged or plain duct ends. 
The holes should be drilled 1/16"—1/8" in diameter or less; the 
holes should not be punched as inwardly projecting jagged 
edges of metal will disturb the air stream. The U-tube ma- 
nometer is connected to each hole in turn by means of a 
thick-walled soft rubber tube and the difference in the height 
of the water columns is read in inches. 


If an elbow intervenes between the hood and the suction- 
measurement location, the pressure loss caused by the elbow 
should be subtracted from the reading to indicate the suction 
produced by the hood and throat alone (see Chapter 5, Figure 
5-14). 


The values for hood entry loss coefficient (h,) for various 
hood shapes are listed in Chapter 5, Figure 5-13. When the 
hood static pressure (SP),) is known, the volumetric flow rate 
can be determined by the following equation: 


SP, 


=1096A |—— "1 
: (1+ he )p 


[9.7] 
where: 
Q= volumetric flow rate, cfm 
A= area of duct attached to hood, ft? 
SP, = U-tube average manometer reading, “wg 
h, = hood entry loss coefficient 
p = actual gas density, lb/m/ft? 


For standard air, Equation 9.7 becomes: 


As noted above, the coefficient of entry, C,, can also be 
used in conjunction with SP, to determine Q. To facilitate 
repeated measurements, it is sometimes convenient to post the 
value of C, directly on the hood near the point where SP,, is 
measured. 


9.4.3 Hood Static Pressure Interpretation: If the hood 
static pressure is known while a system is functioning prop- 
erly, its continued effectiveness can be assured so long as the 
original value is not changed. Any change from the original 
measurement can only indicate a change in velocity in the 
branch and, consequently, a change in volumetric flow 
through the hood. This relationship will be true unless: 1) a 
hood design change has affected the entrance loss; 2) there 
are obstructions or accumulations in the hood or branch ahead 
of the point where the hood static pressure reading was taken; 
or 3) the system has been altered or added to. Depending on 
the location of the obstruction in the duct system, restrictions 
of the cross-sectional area will reduce the air flow although 
hood suction may increase or decrease. 


Pressure readings vary as the square of the velocity or 
volumetric flow rate. To illustrate, an indicated reduction in 
static pressure readings of 30% would reflect a volumetric 
flow rate (or velocity) decrease of 6%. 


A marked reduction in hood static pressure often can be 
traced to one or more of the following conditions: 


1. Reduced performance of the exhaust fan caused by 
reduced shaft speed due to belt slippage, wear, or 
accumulation on rotor or casing that would obstruct air 
flow. 


2. Reduced performance caused by defects in the exhaust 
piping such as an accumulation in branch or main ducts 


9-14 Industrial Ventilation 


-~ 5 IN.= 16 D- crs 


— 


) 
A 


0.312 IN.= 1D 


SECTION A-A 


0.156 IN. RAD. 
8 HOLES — 0.04 IN. DIA. 
EQUALLY SPACED 


FREE FROM BURRS 
NOSE SHALL BE FREE 


FROM NICKS AND BURRS. 


| ; ae NOTE :OTHER SIZES OF PITOT TUBES WHEN REQUIRED, MAY BE BUILT USING THE 
| | al SAME GEOMETRIC PROPORTIONS WITH THE EXCEPTION THAT THE STATIC 


‘oa ORIFICES ON SIZES LARGER THAN STANDARD MAY NOT EXCEED 0.04 IN. IN 


Lt. INNER TUBING ~ APPROX. 


| 0.125 IN 0.0. x 21 B&S GUAGE 


Pee 
We fecal he, eae, Seis 


“TS STATIC. PRESSURE 


'| I~ OUTER TUBING 
71 0.312 IN O.D. x APPROX. 18 B&S GUAGE 


<Q 


p TOTAL PRESSURE 


FIGURE 9-9. Standard Pitot tubes 


due to insufficient conveying velocities, condensation 
of oil or water vapors on duct walls, adhesive charac- 
teristics of material exhausted, or leakage losses 
caused by loose clean-out doors, broken joints, holes 
worn in duct (most frequently in elbows), poor con- 
nection to exhauster inlet, accumulations in ducts or 
on fan blades. 


3. Reduced air flow rate also can be charged to additional 
exhaust duct openings added to the system (sometimes 
systems are designed for future connections and more 
air than required is handled by present branches until 
future connections are made) or change of setting of 
blast gates in branch lines. Blast gates adjust the air 
distribution between the various branches. Tampering 
with the blast gates can seriously affect such distribu- 
tion and therefore they should be locked in place 
immediately after the system has been installed and its 
effectiveness verified. Fan volume contro] dampers 
also should be checked. 


4. Reduced volumetric flow may be caused by increased 
pressure loss through the dust collector due to lack of 
maintenance, improper operation, wear, etc. These 


B 7 “ DIAMETER. THE MINIMUM PITOT TUBE STEM DIAMETER RECOGNIZED 
UNDER THIS CODE SHALL BE 0.10 IN. IN NO CASE SHALL THE STEM 
DIAMETER EXCEED 1/30 OF THE TEST DUCT DIAMETER. 


effects will vary with the collector design. Refer to 
operation and maintenance instructions furnished with 
the collector or consult the equipment manufacturer. 


9.4.4 Velocity Pressure: For measuring velocity pressure 
to determine air velocity, a standard Pitot tube may be used. 
A large volume of research and many applications have been 
devoted to the subject of flow measurements by this instru- 
ment, which was developed by Henry Pitot in 1734 while a 
student in Paris, France. A standard Pitot tube (see Figure 9-9) 
needs no calibration if carefully made and the accuracy of 
velocity pressure readings obtained are considered to be ac- 
curate at velocities above 600 fpm (see Table 9-1). For more 
details concerning specifications and application of the Pitot 
tube, see the "Standard Test Code" published by the American 
Society of Heating, Refrigerating and Air Conditioning Engi- 
neers and the Air Moving and Conditioning Association.©!°) 


The device consists of two concentric tubes — one meas- 
ures the total or impact pressure existing in the air stream; the 
other measures the static pressure only. When the annular 
space and the center tube are connected across a manometer, 
the difference between the total pressure and the static pres- 
sure is indicated on the manometer. This difference is the 


AMnOaOA A AaAaNMno 
ONWOW WN wW wt shoot 
Not Nn + oO MLO 
DON ONMAD i 
0000 Oo Oo000 | 
' ; obo i 
t | | 
( ae 7 it ; t 
_ ° PSS : 
1 j @ t i 
foros / { 
| | ° i 
| { 
, | 9 14 il | 
{ly bo | { 
or -0-@—O-—~@ $ en —|~ @-——o—-0-0-6 
} i 
{ 
\ ’) 
if | 
9 A 
: e 4 i 
~— : oe 


10 POINT PITOT TRAVERSE | 
IN A CIRCULAR DUCT. 


(GREATER THAN 6” DIAM.) | 
10 OR 20 LOCATIONS IN CENTERS | 
OF EQUAL ANGULAR AREA. | 

| 


FIGURE 9-10A. 10-point Pitot traverse in a circular duct 


velocity pressure. 
The velocity pressure can be used to compute the velocity 


of the air stream if the density of the air is known. The 
following equation can be used: 


v =1096 [YP 19.9] 
p 


where: 
VP = velocity pressure, "wg 
p = actual gas density, lb/m/ft? 


Where air is at standard conditions (p = 0.075 Ib/m/ft*), 
Equation 9.9 becomes: 


V = 4005/VP [9.10] 


For example, if the temperature of the air stream varies more 
than 30 F from standard air (70 F and 29.92 "Hg) or the altitude 
of the site is more than 1,000 feet above or below sea level or 
the moisture content of the air is 0.02 Ib/Ib of dry air or greater, 
the actual gas density (p) must be used. 


Velocity pressure versus velocity tables for standard air can 
be found in Chapter 5 (Tables 5-7A and 5-7B). These tables 
can be used for air at densities other than standard conditions 
by determining an equivalent velocity pressure. 

VP, 


VP, =—2 


9.11 
7 (11 


Testing of Ventilation Systems 9-15 


i aa a (a) aa 
i Nox xt oO on | 
| nw o) fon) <x 
i O70) Es NE ev | 
one) Q =) GO 
5 | ; | | 
t H | aan eae | 
| ? Pf 
| | 6 i $ 
/| 
i} | | | |\ 
td Foe ~e@-— ed °. oe ey 
\ 1 | 
i \ 6 / 
< | 
i "es hae | 
6 POINT PITOT TRAVERSE | 
\ IN A CIRCULAR DUCT. 
(6" DIAM. OR LESS) | 
6 OR 12 LOCATIONS IN. CENTERS | 
OF EQUAL ANGULAR AREA. | 


FIGURE 9-108. 6-point Pitot traverse in a circular duct 


where: 
VP, = equivalent velocity pressure, "wg 
VP, = measured velocity pressure, "wg 
df= density factor coefficient 


The equivalent VP then can be used in the velocity pressure 
versus velocity table selected to give the actual velocity at 
duct conditions. 


A number of techniques can be used to determine the 
volumetric flow rate at hood openings and at other points in 
an exhaust system using the fluid flow principles previously 
described. The method selected will depend on the degree of 
accuracy required, time available for testing, and the type of 
test data required. It is extremely important that measurements 
taken at the time of the tests include all necessary information 
to determine the gas density to permit the calculation of the 
actual velocity and volumetric flow rate. 


9.5 PITOT TRAVERSE METHOD 


Because the air flow in the cross-section of a duct is not 
uniform, it is necessary to obtain an average by measuring VP 
at points in a number of equal areas in the cross-section. The 
usual method is to make two traverses across the diameter of 
the duct at right angles to each other. Readings are taken at 
the center of annular rings of equal area (see Figures 9-10A 
and 9-10B). Whenever possible, the traverse should be made 
7 % duct diameters or more downstream from any major air 


9-16 Industrial Ventilation 


{ueg———tetue——-- YP MAX. 
to} [nace 


i= 


—— 


= 


A: FULLY DEVELOPED VP DISTRIBUTION 


B: GOOD VP DISTRIBUTION. (ALSO SATISFACTORY FOR 
FLOW INTO FAN INLETS. BUT MAY BE UNSATIS— 
FACTORY FOR FLOW INTO INLET BOXES — MAY 
PRODUCE SWIRL IN BOXES.) 


P= VP MAX. 


—ee| | nee— 


VP MAX. 
10 


i C: SATISFACTORY VP DISTRIBUTION — MORE THAN 
75% OF VP READINGS GREATER THAN ve MAX. 


D: DO NOT USE! UNSATISFACTORY VP DISTRIBUTION — f 
LESS THAN 75% OF VP READINGS GREATER THAN | § 


VP_ MAX. 
10 


— | unit 


40% 


VP_ MAX. VP MAX. 


I 


20% 


F: DO NOT USE! UNSATISFACTORY VP DISTRIBUTION — : 
LESS THAN 75% OF VP READINGS GREATER THAN) @ 
VP_ MAX. 

10 


E: DO NOT USE! UNSATISFACTORY VP DISTRIBUTION — 
LESS THAN 75% OF VP READINGS GREATER THAN 
VP_ MAX. 
10 


VELOCITY PRESSURE 
DISTRIBUTIONS 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


PITOT TRAVERSE POINTS IN A RECTANGULAR 
DUCT. CENTERS OF 16 TO 64 EQUAL AREAS. 
LOCATIONS NOT MORE THAN 6” APART. 


bs oe ee 


MA 


FIGURE 9-12. Pitot traverse points in a rectangular duct 


disturbance such as an elbow, hood, branch entry, etc. Where 
measurements are made closer to disturbances, the results 
must be considered subject to some doubt and checked against 
a second location. If agreement within 10% of the two tra- 
verses is obtained, reasonable accuracy can be assumed and 
the average of the two readings used. Where the variation 
exceeds 10%, a third location should be selected and two air 
flows in the best agreement averaged and used. The use of a 
single centerline reading for obtaining average velocity is a 
very coarse approximation and is NOT recommended. 


The reason for the uncertainty and variation in measure- 
ments is the non-uniformity of air flow after a disturbance. 
Figure 9-11 shows some air flow patterns that could develop 
after a disturbance and the resulting difficulties in obtaining 
good reliable measures are evident. 


For round ducts 6" and smaller, at least 6 traverse points 
should be used. For round ducts larger than 6" diameter, at 
least 10 traverse points should be employed. The number of 
traverse locations on each diameter and the number of tra- 


TABLE 9-5. Distance from Wall of Round Pipe to Point Reading 
(nearest 1/8 inch) for 6-Point Traverse 


Duct R, R, R; Ry Rs Rg 
DIA | .043 DIA| .146 DIA} .296 DIA| .704 DIA | .854 DIA] .957 DIA 
t 
3 1/8 1/2 718 2 1/8 2718 
3 1/2 1/8 1/2 | 1 2 1/2 
4 1/8 5/8 | 1 1/8 2718 
r is 
41/2 1/4 5/8 13/8 3 1/8 
5 1/4 I 3/4 Ie 1 1/2 3 1/2 
5 1/2 1/4 3/4 1 5/8 37/8 
oe <r es 
1 6 1/4 i 7/8 1 3/4 "| 41/4 | 


Testing of Ventilation Systems 9-17 


verse diameters required are determined by the need for 
accuracy and the symmetry of the measured values. Where 
uniform velocity pressure profiles exist, a single traverse 
along one diameter may be adequate. Where the values are 
moderately skewed, the use of two diameters is recom- 
mended. For greater accuracy, a third diameter should be 
used. Six, ten, and twenty point traverse points for various 
duct diameters are given in Tables 9-5, 9-6, and 9-7. To 
minimize errors, a Pitot tube smaller than the standard 5/16" 
O.D. should be used in ducts less than 12" in diameter. 


For square or rectangular ducts, the procedure is to divide 
the cross-section into a number of equal rectangular areas and 
measure the velocity pressure at the center of each. The 
number of readings should not be less than 16. However, 
enough readings should be made so the greatest distance 
between centers is approximately 6" (see Figure 9-12). 


The following data are essential and more detailed datamay 
be taken if desired: 


e The area of the duct at the traverse location. 
e Velocity pressure at each point in the traverse. 


° Temperature of the air stream at the time and location 
of the traverse. 


The velocity pressure readings obtained are converted to 
velocities and the velocities, not the velocity pressures, are 
averaged. Where more convenient, the square root of each of 
the velocity pressures may be averaged and this value then 
converted to velocity (average). The measured air flow is then 
the average velocity multiplied by the cross-sectional area of 
the duct (Q = VA). Where conditions are not standard, see 
"Corrections for Non-Standard Conditions." 


The Pitot tube cannot be used for measuring velocities less 
than 600 fpm in the field. It is susceptible to plugging in air 
streams with heavy dust and/or moisture loadings. A vibration 
free mounting is necessary if using a liquid manometer. See 
Reference 9.10 for special instrumentation which can be used 
to measure low velocities. 


Modified Pitot Tubes: Modified Pitot tubes have been 
made in an effort to reduce plugging difficulties encountered 
in heavy dust streams or to increase manometer differentials 
enabling the measurement of lower velocities in the field. 
These are referred to as "S"-type (Staubscheide) tubes, They 
usually take the form of two relatively large impact openings, 
one facing upstream and the other facing downstream. Such 
tubes are useful when thick-walled ducts, such as boiler 
stacks, make it difficult or impossible to insert a conventional 
Pitot tube through any reasonably sized port opening. They 
require only initial calibration for all conditions. 


Measurements made with an "S"- type Pitot tube cannot be 
used directly. The tube first must be calibrated against a 
standard Pitot tube and the velocity pressure measured cor- 
rected to the actual velocity pressure. 


9-18 Industrial Ventilation 


TABLE 9-6. Distance from Wall of Round Pipe to Point of Reading (nearest 1/8 inch) for 10-Point Traverse 


R, R, R3 Ry Rs Rg Ry Rg Rg Rio 
DUCT DIA | 0.026DIA | 0.082DIA clas 0.226DIA | 0.342DIA | 0.658DIA |_0.774DIA 0.854DIA | 0.918DIA | O.974DIA 

4 118 3/8 5/8 718 2 5/8 3 1/8 3 3/8 3 5/8 3 7/8 
41/2 1/8 3/8 5/8 1 3 3 1/2 37/8 41/8 43/8 

5 1/8 | 3/8 | 3/4 11/8 3 1/4 i 37/8 oo 1/4 4 8 | 47/8 | 
5 1/2 | 1/8 4/2 3/4 1 1/4 3 5/8 41/4 4 3/4 5 5 3/8 

6 1/8 1/2 718 13/8 4 tat 5/8 he 5 1/8 5 1/2 5 7/8 

7 1/8 5/8 1 15/8 45/8 5 3/8 6 6 3/8 6 7/8 

8 1/4 5/8 | 1 1/8 1 3/4 5 1/4 6 1/4 bs 6 7/8 7 3/8 7 3/4 

9 1/4 3/4 11/4 2 57/8 7 7 3/4 | 8 1/4 8 3/4 

10 [ 1/4 718 | 1 1/2 2 1/4 6 5/8 7 3/4 8 1/2 9 1/8 9 3/4 

1 1/4 | 7/8 1 cel 2 1/2 7 114 8 1/2 12 938 | 10 ed 10 3/4 

12 | 3/8 { 1 3/4 2 3/4 77/8 9 1/4 10 1/4 11 11 5/8 

13 i) 3/8 { I 17/8 27/8 8 1/2 10 1/8 11 1/8 12 12 8 

14 ae 3/8 11/8 2 ast 3 1/8 9 1/4 10 7/8 12 12 7/8 13 5/8 

15 3/8 11/4 2 1/4 3 3/8 97/8 11 5/8 12 3/4 13 3/4 14 5/8 


16 3/8 1 1/4 2 3/8 3 5/8 5 1/2 10 1/2 12 3/8 13 5/8 14 3/4 155/8 | 


17 1/2 13/8 2 1/2 37/8 5 3/4 11 1/4 13 1/8 14 1/2 15 5/8 16 1/2 
18 1/2 11/2 2 5/8 41/8 6 1/8 11718 13 7/8 15 3/8 16 1/2 17 112 


19 112 1112 23/4 41/4 612 | 1212 | 1434 | 1614 | 1742 | 1891/2 

20 112 15/8 27/8 41/2 1318 | 1512 | 1718 | 1838 | 4191/2 
22 5/8 13/4 31/4 5 712 | 14112 17 18314 | 2014 | 2138 
24 | 518 2 31/2 51/2 814 | 1534 | 1812 | 20112 22 23 318 


26 5/8 2 1/8 3 3/4 57/8 87/8 17 118 20 1/8 2214 | 237/18 |  253/8 
28 3/4 21/4 4118 | 6 3/8 9 5/8 18 3/8 21 5/8 23 7/8 25 3/4 27 1/4 
30 3/4 au 2 1/2 43/8 6 3/4 10 1/4 19 3/4 23 1/4 25 5/8 27 12 29 1/4 
32 718 25/8 45/8 7 1/4 "1 21 24 3/4 27 3/8 29 3/8 31 1/8 
34 718 2 3/4 5 73/4 11 5/8 22 3/8 26 1/4 i 29 31 1/4 | 33 1/8 
36 1 3 5 1/4 8 1/8 12 3/8 23 5/8 27 718 30 3/4 33 35 

38 | 1 3 1/8 | 5 1/2 8 5/8 13 25 29 3/8 32 1/2 34 7/8 za 37 

40 | 1 31/4 57/8 | 9 13 5/8 26 3/8 31 34 1/8 36 3/4 39 

42 1118 33/8 6 1/8 91/2 14 3/8 27 5/8 32 1/2 35 7/8 38 5/8 40 718 
44 11/8 4 3 5/8 6 3/8 10 15 29 34 37 5/8 _| 40 3/8 42718 
46 i 11/4 33/4 | 6 3/4 10 3/8 15 3/4 30 1/4 35 5/8 38 1/4 42 1/4 44 3/4 


48 11/4 4 7 10 7/8 16 3/8 31 5/8 a7 18 | AN 44 46 3/4 


Testing of Ventilation Systems 9-19 


TABLE 9-7. Distance from Wall of Round Pipe to Point of Reading (nearest 1/8 inch) for 20 Point Traverse 


fis 
, J 07960 | 08350 | 08710 
1 1/2 2 5/8 3 7/8 5 1/8 6 5/8 8 1/8 10 12 1/4 
24 1/2 27 3/4 30 31 7/8 33 3/8 34 7/8 36 1/8 37 3/8 38 1/2 
1/2 1 5/8 2 7/8 4 1/8 5 3/8 6 7/8 8 5/8 10 1/2 12 7/8 
25 3/4 29 1/8 31 1/2 33 3/8 35 1/8 36 5/8 37 7/8 39 1/8 40 3/8 
1/2 1 3/4 3 4 1/4 5 5/8 7 1/4 9 11 13,°1/2 
26 7/8 30 1/2 33 35 36 3/4 38 3/8 39 3/4 41 42 1/4 
5/8 1 3/4 3.1/8 4 1/2 6 7 5/8 9 3/8 11 1/2 14 1/8 
28 1/8 31 7/8 34 1/2 36 5/8 38 3/8 40 41 1/2 42 7/8 Kh 1/4 
5/8 1 7/8 3 1/4 4 5/8 6 1/4 7 7/8 9 3/4 12 14 3/% 
29 3/8 33 1/4 36 38 1/4 40 1/8 41 3/4 43 3/8 uh 3/4 46 1/8 
5/8 2 3 3/8 4 7/8 6 1/2 8 1/4 10 1/4 12 1/2 15 3/8 
30 5/8 34 5/8 37 1/2 39 3/4 41 3/4 43 1/2 45 1/8 46 5/8 48 
5/8 2 3.1/2 5 6 3/4 8 1/2 10 5/8 13 15 7/8 
31 7/8 36 1/8 39 41 3/8 43 1/2 45 1/4 47 48 1/2 
5/8 21/8 3 5/8 5 1/4 7 8 7/8 13 1/2 
33 37 1/2 40 1/2 43 45 1/8 47 4 50 3/8 


3/4 
34 1/4 45 / 


39 1/2 


16 1/4 
41 1/2 


17 (1/8 
43 1/2 


17 7/8 
45 3/8 


18 5/8 
47 3/8 


19 3/8 
49 3/8 


20 1/8 


ve uw 
eo o 
Ne 

~ 
or 


~TE™N 
oo 


cr 
om 
w 
™ 
wn 
we 
Ww 
~ 
) 


22 1/2 
57 1/4 


3/4 2 1/4 3 7/8 5 5/8 7 1/2 9 1/2 11 7/8 14 1/2 17 3/4 
35 1/2 &O 1/4 43 1/2 &6 1/8 48 1/2 50 1/2 52 3/8 54 1/8 55 3/4 
3/4 2 3/8 4 5 7/8 7 3/4 9 7/8 12 1/4 15 18 3/8 
36 3/4 41 5/8 45 &7 3/4 50 1/8 52 1/h 54 1/8 56 57 5/8 
3/4 2 3/8 4 1/8 6 8 10 1/% 12 5/8 15 1/2 19 
37 7/8 43 46 1/2 49 3/8 51 3/4 54 56 57 7/8 59 5/8 
3/4 21/2 4 l/h 6 1/4 8 1/4 10 1/2 13 1/8 16 19 5/8 
39 1/8 4h 3/8 48 50 7/8 53 1/2 55 3/4 57 3/4 59 3/4 61 1/2 
7/8 2 5/8 4 3/8 6 3/8 8 1/2 10 7/8 1301/2 16 1/2 20 1/4 
40 3/8 4&5 3/% 49 1/2 52 1/2 55 1/8 57 1/2 59 5/8 61 5/8 63 3/8 
7/8 2 5/8 4 1/2 6 5/8 8 3/4 11 1/4 13 7/8 17 20 7/8 
41 5/8 &7 1/8 51 54 1/8 56 3/4 59 1/4 61 3/8 63 1/2 65 3/8 


23 1/4 
59 1/4 


24 1/8 
51 1/4 


24 7/8 
63 1/4 


25 5/8 
65 1/8 


26 3/8 
67 1/8 


21 1/2 
67 1/4 


27 1/8 
69 1/8 


7/8 2 3/4 & 3/4 6 3/4 9 11 °1/2 14 1/4 
42 7/8 48 1/2 52 1/2 55 3/% 58 1/2 61 63 1/4 
7/8 2 3/4 4 7/8 7 9 1/4 11 7/8 14 3/4 
uy 50 54 57 1/% 60 1/8 62 3/4 65 69 1/4 
2 7/8 5 71/8 9 1/2 12 1/8 18 1/2 22 5/8 
51 3/8 55 1/2 58 7/8 61 7/8 64 1/2 69 71 «1/8 
1 3 5 1/8 7 3/8 9 7/8 12 1/2 15 1/2 19 23 1/4 
46 1/2 52 3/4 57 60 1/2 63 1/2 66 1/8 68 5/8 70 7/8 73 
3 5 1/4 71/2 10 1/8 12 7/8 15 7/8 19 1/2 23 7/8 
3/4 5&4 1/8 58 1/2 62 1/8 65 1/8 67 7/8 70 1/2 72 3/4 75 
3.1/8 5 3/8 7 3/4 10 3/8 13 1/8 16 3/8 20 24 1/2 31 
55 1/2 60 63 5/8 66 7/8 69 5/8 72 1/4 74 5/8 76 7/8 79 


Lad 
i) 


28 
71 1/8 


28 3/4 
73°1/8 


w 
ite) 
~~ 
oa.) 
Om 
Ou 
~e 
~~ 
© © 


29 1/2 


30 1/4 


e - 
oe Se 


~~ ~ Wn 
~~ w ue 
Lanth“al 
~ns 
re 


9-20 Industrial Ventilation 


6 
5 
4 
3 
2 
/ 

7] 


2-022" 
FIGURE 9-13, U-tube manometer 


Other modified forms of the Pitot tube are the air foil 
pitometer, the Pitot venturi, and the air speed nozzle, to name 
a few.0.7.°8) 


Pressure Sensors: Pressure sensors can be used in conjunc- 
tion with the pitot tube to measure pressures existing within 
ventilation systems, These devices are described below. 


U-Tube Manometer: The vertical U-tube (see Figure 9-13) 
is the simplest type of pressure gauge. Usually calibrated in 
inches water gauge, it is used with various fluid media such 
as alcohol, mercury, oil, water, kerosene and special manome- 
ter fluids. The U-tube may be used for either portable or 
stationary applications. Available commercial units offer a 
wide latitude in range, number of columns, and styles. Tubes 
are usually of all-plastic construction to minimize breakage. 
One leg may be replaced by a reservoir or well (well-type 
manometer) with the advantage of easier manometer reading. 


Gea 


QA A&A UHM ~ DN Hh HW A ® 


/+#/=2" 


S 


a“ 


O9~hHWAAAN ® © 
Lenya 


6-4 =2" 


Inclined Manometer (Figure 9-14): Increased sensitivity 
and scale magnification is realized by tilting one leg of the 
U-tube to form an inclined manometer or draft gauge. The 
inclined manometer gives increased accuracy and permits 
lower readings. In commercial versions, only one tube of the 
small bore is used and the other leg is replaced by a reservoir. 
The accuracy of the gauge is dependent on the slope of the 
tubes. Consequently, the base of the gauge must be leveled 
carefully and the mounting must be firm enough to permit 
accurate leveling. The better draft gauges are equipped with 
a built-in level, leveling adjustment and, in addition, a means 
of adjusting the scale to zero. Some models include over-pres- 
sure safety traps to prevent loss of fluid in the event of pressure 
surges beyond the manometer range. 


A modification of the inclined manometer is the inclined- 
vertical gauge in which the indicator leg is bent or shaped to 
give both a vertical and inclined portion — the advantage is 


FIGURE 9-14. Inclined manometer 


FIGURE 9-15. Aneroid gauge 


smaller physical size for a given range while retaining the 
refined measurement afforded by the inclined manometer. As 
in the U-tube and inclined gauges, the commercial units 
available offer a wide choice in range, number of columns, 
and calibration units. 


Aneroid Gauges: This type of gauge is used as a field 
instrument in ventilation studies for measuring static, veloc- 
ity, or total pressure with a Pitot tube or for single tube static 
pressure measurements. A number of manufacturers offer 
gauges suitable for the measurement of the low pressures 
encountered in ventilation studies. Perhaps the best known of 
this type is the Magnehelic™ gauge (Figure 9-15). The prin- 
cipal advantages of this gauge can be listed as follows: easy 
to read, greater response than manometer types; very portable 
— small physical size and weight; absence of fluid means less 
maintenance; and mounting and use in any position is possible 
without loss of accuracy. Principal disadvantages are that the 
gauge is subject to mechanical failure, requires periodic cali- 
bration checks, and occasional recalibration. 


Electronic Aneroid Gauges: Commercial instruments are 
now available which will measure and record static pressure 
as well as integrate velocity pressure directly to velocity using 
the pressure sensing principles of an aneroid gauge. This type 
of instrument can be connected directly to a standard Pitot 
tube and used in the same manner as a U-tube manometer. 
The instruments are light in weight, easily hand-held, and can 
be equipped with an electronic digital display or print recorder 
with measurement data in either English or S.I. units. Because 
they are battery powered, periodic servicing is required as is 
calibration. 


Testing of Ventilation Systems 9-21 


9.6 CORRECTIONS FOR NON-STANDARD CONDITIONS 


Air velocities sometimes are measured at conditions sig- 
nificantly different from standard. If these conditions are 
ignored, serious errors can be introduced in the determination 
of the actual duct velocity and the volumetric flow rate(s) in 
the system. Elevation, pressure, temperature, and moisture 
content all affect the density of the air stream. The actual 
density present in the system must be used in either Equation 
9.2 or 9.9 to determine the actual velocity. 


Correction for changes in elevation, duct pressure, and 
temperature can be made independently of each other with 
reasonable accuracy. The individual correction coefficients 
are multiplied together to determine the change from standard 
air density. The actual air density becomes: 


p = 0.075df and [9.12] 
df = CF, CF, CF, 9.13] 
where: 
CF, = correction for elevations outside the range of 
+ 1000 ft 
CF, = correction for local duct pressures greater than 
+ 20 "wg 
CF,= correction for temperatures outside the range of 
40 to 100 F 


One exception to this general rule is when elevations sig- 
nificantly different from sea level are coupled with high 
moisture content. Where this occurs, a psychrometric chart 
based upon the barometric pressure existing at the elevation 
of concern should be used. See Chapter 5 for an explanation 
of the determination of density using a psychrometric chart 
when moisture content and temperature are significantly dif- 
ferent from standard. 


The correction coefficient for elevation, CF,, can be given by 


CF, = [1-(6.73x 10% y(z)]>?*° [9.14] 
where: 
= elevation, ft. 


The correction coefficient for local duct pressure, CF, can 
be given by 


Geese [9.15] 
407 
where: 


SP = static pressure, "wg. (Note that the algebraic 
sign of SP is important.) 


The correction coefficient for temperature, CF,, can be 
given by 


530 
= [9.16] 
t” t+460 


9-22 Industrial Ventilation 


where: 
t= dry-bulb temperature, F (Note: Algebraic sign 
of t must be used) 


Density factors (df) for various altitudes, barometric pres- 
sures, and temperature conditions are shown on Table 9-8. 

Example 1: A velocity pressure reading of 1.0 “wg was 
taken with a Pitot tube in a duct where the dry-bulb tempera- 
ture is 300 F, the moisture content is negligible and the static 
pressure is - 23.5 "wg. The system is installed at an elevation 
of 5000 feet. What would the density and actual velocity be 
at that point? 


As the moisture content is unimportant, Equations 9.12 and 
9.13 can be used directly to determine the density. 


The individual correction coefficients can be found from 
Equations 9.14 through 9.16 as 


CF, = [1-(6.73 x 10° )(5000)}°® = 0.84 


_ 407-235 _ 


CF, 0.94 
407 
CF, =" = 0.70 
300+ 460 


Then the density at this condition would be 


p = (0.075)(0.84)(0.94)(0.70) = 0.0415 Ibm/ ft 
and the velocity from Equation 9.9 would be 
10 
0.0415 


Note that an error of 26% would result if standard density had 
been assumed. 


V = 1096 


= 5380 fpm 


Example 2: A swinging vane anemometer is used to 
determine the velocity in a duct at sea level where the dry bulb 
temperature is 250 F, the SP = -10 "wg and moisture is 
negligible. What is the actual duct velocity if the anemometer 
reading is 3150? 


The temperature correction coefficient is 0.75 from Equa- 
tion 9.16 and the density would be 


p = (0.75)(0.075) = 0.0563 Ibm/ ft® 
Therefore, the actual velocity in the duct would be 
Voorrected es Vmneasuredv Of 


0.075 
0.0563 


V, = 3150 


corrected — = (3150)(115) = 3636 fpm 

9.6.1 Example Traverse Calculations: Measurement of 
air velocity at non-standard conditions requires calculation of 
the true air velocity, accounting for difference in air density 
due to air temperature, humidity, and barometric pressure. 


The following calculations illustrate the method of calculation 
and the effect of varying air density. 


1. Standard Conditions: 


Air Temp. = 79 F; Wet-Bulb Temp. = 50 F 
Barometer = Std. (29.92 "Hg); 24" Duct Diameter 


Pitot Traverse #2 
Pitot Traverse #1 (1 to Traverse #1) 
Traverse Pt. VPy Vv," Traverse Pt.  VPy V,* 
1 0.22 1879 1 0.23 1921 
2 0.28 2119 2 0.27 2081 
3 0.32 2260 3 0.33 2301 
4 0.33 2301 4 0.34 2335 
5 0.34 2335 5 0.34 2335 
6 0.35 2369 6 0.35 2369 
7 0.33 2301 7 0.34 2335 
8 0.32 2230 8 0.32 2260 
9 0.30 2193 9 0.32 2230 
10 0.24 1962 10 0.25 2003 
21949 22170 


*Calculated from Equation 9.9 or Chapter 5, Table 5-7 


21949+22170 44119 


AverageVelocity, V, = 50 Sane 


= 2205.9 = 2206 fpm 


Q, = VA = 2206 x 3.142 = 69312 = 6931scfm 


2. Elevated Temperature: 


Air Temp. = 150 F; Wet-Bulb Temp. = 80 F 
Barometer = Std.; 24" Outside Diameter Duct 


Pitot Traverse #2 
Pitot Traverse #1 (1 to Traverse #1) 
Traverse Pt. VPy V,* Traverse Pt. VPy Vv," 

1 0.22 2015 1 0.23 2060 
2 0.28 2275 2 0.27 2235 
3 0.32 2430 3 0.33 2465 
4 0.33 2470 4 0.34 2505 
5 0.34 2505 5 0.34 2505 
6 0.35 2540 6 0.35 2540 
7 0.33 2470 7 0.34 2905 
8 0.32 2395 8 0.32 2430 
9 0.30 2355 9 0.32 2395 
10 0.24 2105 10 0.25 2150 
23560 23790 


*Calculated from Equation 9.9 or Chapter 5, Table 5-7 


To determine the air velocity at standard conditions (V,) 
for each VPy, the density (p) can be calculated using Equa- 
tions 9.12 and 9.16: 


p = 0.075 x (530 = 610) = 0.065lbm/ ft 


Using Equation 9.11, each VPy is multiplied by the ratio 
0.075 + 0.065 and the resulting V, values averaged. 


23560+23790 47350 
20 20 


Average Velocity, V, = 


= 2368 fpm 


Q, = VA = 2368 x 3.142 = 7440.3 = 7440 scfm 
Short Method: 


Find: “standard velocity” average from measured VP, = 2206 
fpm (From #1) 


VP for 2206 fpm = 0.30 (Equation 9.10); at 150 F, 
density = 0.075 + 0.87 = 0.065 Ib/m/ft® 


VPs = VPm x (0.075 = 0.065) = 0.346 = 0.35 “wg 
Vs = 2370 fpm 


Lo 


. Elevated Temperature and Moisture: 


Air Temp = 150 F; Wet-Bulb Temp. = 140 F 
Barometer = Std.; 24" Outside Diameter Duct 


Pitot Traverse #2 
Pitot Traverse #1 (1 to Traverse #1) 
Traverse Pt. VPy V,* Traverse Pt. VPu V,* 

1 0.22 2100 1 0.23. ©2145 
2 0.28 2370 2 0.27 = 2325 
3 0.32 2530 3 0.33 2570 
4 0.33 2570 4 0.34 =. 2610 
5 0.34 2610 5 0.34 ~—.2610 
6 0.35 2645 6 0.35 2645 
7 0.33 2570 7 0.34 2610 
8 0.32 2490 8 0.32 2530 
9 0.30 2450 9 0.32 2490 
10 0.24 2190 10 0.25 = 2235 
24525 24770 


*Calculated from Equation 9.9 or Chapter 5, Table 5-7 


To determine the air velocity at standard conditions (V,) 
for each VPw, the air density (p) can be calculated using the 
psychrometric charts found in Chapter 5. 


p = 0.075 x 0.80 (density coefficient — mixture) 


= 0.06lb/m/ ft® 


Testing of Ventilation Systems 9-23 


Using Equation 9.11, each VPy is multiplied by the ratio 
0.075 + 0.06 and the resulting V, values averaged. 


_ 24525 + 24770 


Average velocity, V 
g y, Vs 50 


= 2460 


49295 
20 
= 2465 fom 
(V, may be found also by the Short Method found in #2.) 
Qactual = VsA = 2465 x 3.142 
= 7745 cfmof air and water mixture 
Weight of mixture = Q, x 0.075 x d 


= 7745 x 0.075 x 0.80 = 465 Ib 


From psychrometric charts, weight of water in mixture = 
0.15 Ib. HyO/lb dry air. 


Weight of dry air 


7 (weight of mixture) 
(weight of dry air + moisture per Ibmdry air) 


465 
= 115 = 404ibm 


404 |bm/nin 404 


Alternate Method: 


From Chapter 5, humid volume = 19.3 ft? of mixture/Ib dry 
air (Interpolate). 


Q, _ 7745 


Weight of dry air = = —— = 403 Ib 
De 2 Bead GG 
-= gos m = 5373 scfm 
0.075 


4. High or Low Altitudes: 


Q,= V, x A where V, can be obtained from Equations 
9.2 and 9.9 in conjunction with Table 9.8 or Equations 
9.12 and 9.13. 


9.7 CHECK-OUT PROCEDURES 


The following procedure may be used on systems (see 
Figure 9-16) that were designed to balance without the aid of 
blast gates. It is intended as an initial verification of the design 
computations and contractors’ construction in new systems, 
but it may be used also for existing systems when design 
calculations are available or can be recomputed. It does not 
detect poor choices of design criteria such as low conveying 
or capture velocities, and consequently, will not reveal inade- 


9-24 


Industrial Ventilation 


TABLE 9-3. Air Density Correction Factor, df 


ALTITUDE RELATIVE TO SEA LEVEL, ft 


—5000 -- 4000 — 3000 — 2000 — 1000 0 


1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 


Barometric Pressure 


35.74 34.51 33.31 32.15 


469.97 453.67 437.84 


31.02 


29.92 28.86 27.82 26.82 25.84 24.89 23.98 23.09 22.22 21.39 20.57 


422.45 407.50 392.98 378.89 365.21 351.93 339.04 326.54 314.42 302.66 291.26 280.21 


Density Factor, df 


1.22 
1.11 
1.02 
0.96 
0.91 
0.84 
0.77 
0.72 
0.67 
0.63 
0.59 
0.56 
0.53 
0.51 
0.48 
0.44 
0.41 
0.38 
0.35 


quate control due to this type of error. Agreement with design 
within + 10% is considered acceptable. 


1. 


nN 


Ww 


Determine volumetric flow in duct with a pitot tr- 
averse. If volumetric flow matches design, go to Step 
4; otherwise, continue with la. 


a. Check fan size against plan; 


b. Check fan speed and direction of rotation against 
design; 


c. Check fan inlet and outlet configuration against plan. 


. Ifa discrepancy is found and corrected, return to Step 


1. If not, measure fan inlet and outlet static pressures 
and compute the fan static pressure. Using fan table, 
check flow, fan static pressure and fan speed (RPM). 
If agreement is acceptable although at some other 
operating point than specified, fan is satisfactory, and 
trouble is elsewhere in the system. Go to Step 3. 


. If fan inlet static pressure is greater (more negative) 


than calculated in the design, proceed to Step 4. If fan 
outlet static pressure is greater (more positive) than 
design, proceed to Step 8. 


. Measure hood static pressure on each hood and check 


against design. If correct, go to Step 10; otherwise, 
continue with Step 4a. 


a. Check size and design of hoods and slots against 
plan; 


1.17 
1.07 
0.99 
0.93 
0.88 
0.81 
0.75 
0.69 
0.65 
0.61 
0.57 
0.54 
0.51 
0.49 
0.46 
0.42 
0.39 
0.36 
0.34 


1.13 
1.03 
0.95 
0.90 
0.85 
0.78 
0.72 
0.67 
0.62 
0.59 
0.55 
0.52 
0.49 
0.47 
0.45 
0.41 
0.38 
0.35 
0.33 


wn 


Co 


b. Examine each hood for obstructions. 


. After all hood construction errors and obstructions 


have been corrected, if hood static pressures are cor- 
rect, return to Step 1; if too low, proceed to Step 6. 


. Measure static pressure at various junctions in ducts 


and compare with design calculations. If too high at a 
junction, proceed upstream until static pressures are 
too low and isolate the trouble. In an area where the 
loss exceeds design: 


a. Check angle of entries to junctions against plan; 
b. Check radii of elbows against plan; 

c. Check duct diameters against plan; 

d. Check duct for obstructions. 


If the static pressure is too low, proceed downstream 
and locate the trouble. 


. After correcting all construction details which deviate 


from specifications, return to Step 1. 


. Measure pressure differential across air cleaning de- 


vice and check against manufacturer's data. If loss is 
excessive, make necessary corrections and return to 
Step 1. If loss is less than anticipated, proceed to Step 
8a. 


a. Check ducts, elbows, and entries as in Step 6a and 
6d. 


A 
ae 


Testing of Ventilation Systems 


9-25 


MEASUREMENT _ 


LOCATION. “OF MEASUREMENT _ _ 


MEASUREMENT USE 


“HOOD STATIC 
PRESSURE 


DISTANCE FROM HOOD — 
3 PIPE #’°S-FLANGED OR PLAIN HOOD 


1 @-TAPERED HOOD 


1. ESTIMATE FLOW:Q= 4005CeA VSP), 
2. CHECK POINT FOR HOOD AND 
SYSTEM PROFORMANCE. 


VELOCITY AND 
STATIC PRESSURE 


BRANCH AND MAINS~PREFERABLY 7.5 


O'S STRAIGHT RUN DOWNSTREAM FROM | 


NEAREST AIR DISTURBANCE 
( EL,ENTRY, ETC..) 


1. TRANSPORT VELOCITY 
2. EXHAUST VOLUME: Q=VA 
3. SP AS SYSTEM CHECK POINT 


CENTERLINE VP 


ISMALL DUCTS LOCATION AS ABOVE. 


CENTERLINE VELOCITY READING ONLY. 


ROUND DUCT ONLY. USE ON SMALL 
DUCTS WHERE TRANSVERSE 
IMPRACTICAL OR WHERE APPROXIMATE 
VOLUME WANTED. 


1 STATIC VELOCITY 
AND TOTAL 
| PRESSURES 


INLET AND OUTLET OF FAN-ANY TWO 
OF THREE READINGS AT EACH 


LOCATION 


1. FAN STATIC AND TOTAL PRESSURES 
FSP= SP, -— SP, — VP, 
TP= SR - SP, + VR - VAR 

2. MOTOR SIZE OR GFM ESTIMATE 


6356 x ME OF FA 
_ SP AS SYSTEM CHECK POINT 


‘STATIC PRESSURE 


INLET AND OUTLET OF COLLECTOR 
DIFFERENTIAL PRESSURE 


“1. COMPARE PRESSURE DROP 


“WITH 
NORMAL OPERATING RANGE 

. CHECKPOINTS FOR MAINTAI 
READINGS ABOVE OR BELOW 
NORMAL INDICATE PLUGGING, WEAR 
OR DAMAGE TO COLLECTOR 
ELEMENTS, NEED OF CLEANI 


ENCE. 


@ = OBSERVATION OF AIR FLOWS SURROUNDING EXHAUST OPENINGS MAY BE VISUALLY AUGMENTED BY 
USE OF SMOKE GENERATORS, TRAILS, AND STREAMERS. 


AMERICAN CONFERENCE 
GOVERNMENTAL 


B ee oT a HYGIENISTS 


OF 


SAMPLE 


SYSTEM 


9216. 


9-26 Industrial Ventilation 


PLANT 
OPERATION EXHAUSTED 


DEPT. 


LINE SKETCH SHOWING POINTS OF MEASUREMENT 


DATE SYSTEM INSTALLED 


HOOD AND TRANSPORT VELOCITY 
T 


DUCT vP SP 
A_|IN. Hy QIN. Hy 
) 


POINT REMARKS 


| 


PITOT TRAVERSE 
PITOT READINGS— SEE TABLES 9-1 TO 9-4 


POINTS 


FAN SP___. (SEE SECTION 6) 
MOTOR 
NAME SIZE 
HP. I 
COLLECTOR 
TYPE & SIZE 


POINT] DIA. | SP a SP 
{INLET 


OUTLET 


+. 


I 


TOTAL VEL. 
AVERAGE VEL. 
CFM 


NOTES 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
j INDUSTRIAL HYGIENISTS 


SURVEY FORM 


b. Check system discharge type and dimensions 
against plans. 


9. If errors are found, correct, and return to Step 1. If no 
errors can be detected, recheck design against plan, 
recalculate, and return to Step 1 with new expected 
design parameters. 


10. Measure control velocities at all hoods where possible. 
If control is inadequate, redesign, or modify hood. 


11. The above process should be repeated until all defects 
are corrected and hood static pressures and control 
velocities are in reasonable agreement with design. 
The actual hood static pressures should then be re- 
corded for use in periodic system checks. A file should 
be prepared containing the following documents: 


¢ System plan 

¢ Design calculations 

e Fan rating table 

¢ Hood static pressures after field measurement 
¢ Maintenance schedule 

* Periodic hood static pressure measurement log 


e Periodic maintenance log 


9.7.1 Difficulties Encountered in Field Measurement: 
The general procedures and instrumentation for the measure- 
ment of air flow have been previously discussed. However, 
special problems connected with air flow necessitate a some- 
what more detailed discussion. 


Some of these special problems are as follows: 


1. Measurement of air flow in highly contaminated air 
which may contain corrosive gases, dusts, fumes, 
mists, or products of combustion. 


2. Measurement of air flow at high temperatures. 


3. Measurement of air flow in high concentrations of 
water vapor and mist. 


4. Measurement of air flow where the velocity is very low 
(see Reference 9.11). 


5. Measurement of air flow in locations of turbulence and 
non-uniform air flow; e.g., discharge of cupolas, loca- 
tions near bends, enlargements or discharges from 
exhaust fans. 


6. Measurement of air flow in connection with isokinetic 
sampling when the velocity is constantly changing. 


Selection of Instruments: The selection of the proper 
instrument will depend on the range of air flow to which the 
instrument is sensitive; its vulnerability to high temperatures, 
corrosive gases, and contaminated atmospheres; its portabil- 
ity and ruggedness and the size of the measuring probe relative 


Testing of Ventilation Systems 9-27 


to the available sampling port. A brief summary of the char- 
acteristics of a few of the instruments which can be used is 
given in Table 9.1. 


In many cases, conditions for air flow measurement are so 
severe that it is difficult to select an instrument. Generally 
speaking, the Pitot tube is the most serviceable instrument; it 
has no moving parts, is rugged, and will stand high tempera- 
tures and corrosive atmospheres when it is made of stainless 
steel. It is subject to plugging, however, when it is used in a 
dusty atmosphere. It cannot be used for measurement of low 
velocities. A special design of Pitot tube can be used for dusty 
atmospheres. In many cases, it is difficult to set up an inclined 
manometer in the field because many readings are made from 
ladders, scaffolds and difficult places. This greatly limits the 
lower range of the Pitot tube. A mechanical gauge can be used 
in place of a manometer. A mechanical gauge is estimated to 
be accurate to 0.02 "wg with proper calibration. 


For lower velocities, the swinging-vane anemometer pre- 
viously described can be used if conditions are not too severe. 
The instrument can be purchased with a special dust filter 
which allows its use in light dust loadings. It can be used in 
temperatures up to 100 F if the jet is exposed to the high 
temperature gases only for a very short period of time (30 
seconds or less). It cannot be used in corrosive gases. If the 
very low velocity jet is used, a hole over 1" in diameter must 
be cut into the duct or stack. 


For very low velocities, anemometers utilizing the heated 
thermocouple principle can be used under special conditions. In 
most cases, these anemometers cannot be used in temperatures 
above 300 F. Contact the manufacturer to determine to what 
degree the thermocouple probe will withstand corrosive gases. 


In sampling work where a match of velocities in the sam- 
pling nozzle and air stream under changing velocities is 
required, the null method is sometimes used. This method 
uses two static tubes or inverted impact tubes, one located 
within the sampling nozzle and the other in the air stream. 
Each is connected to a leg of the manometer; the sampling 
rate is adjusted until the manometer reading is zero. 


REFERENCES: 


9.1 American Society of Heating, Refrigerating and Air 
Conditioning Engineers: 1985 Fundamentals Volume. 
ASHRAE, Atlanta, GA (1985). 


9.2 Air Moving and Control Association, Inc.: Field Per- 
formance Measurements, Publication 203-81. AMCA, 
Arlington Heights, IL (1981). 


9.3 Brandt, A.D.: Industrial Health Engineering, John 
Wiley and Sons, New York (1947). 


9.4 Air Movement and Control Association, Inc.: Test 
Code for Air Moving Devices, AMCA Standard 210- 
86. AMCA, Arlington Heights, IL (1986). 


9.5 Farant, J.P; McKinnon, D.L.; McKenna, T.A.: Tracer 


9-28 Industrial Ventilation 


Gases as a Ventilation Tool: Methods and Instrumen- 
tation. In: Ventilation '85—Proceedings of the First 
International Symposium on Ventilation for Contami- 
nant Control, pp. 263-274. Elsevier Press, Amsterdam, 
The Netherlands (1986). 


9.6 American Society of Mechanical Engineers: Power 
Test Codes, Chapter 4, Flow Measurement. ASME, 
Lubbock, TX (1959). 


9.7 First, M.D.; Silverman, L.: Airfoil Pitometer, Indus- 
trial and Engineering Chemistry, 42:301-308 (Febru- 
ary 1950). 


9.8 American Society of Mechanical Engineers: Fluid 


Meters —-Their Theory and Applications. ASME, 
Lubbock, TX (1959). 


9.9 Hama, G.: A Calibrating Wind Tunnel for Air Meas- 
uring Instruments. Air Engr. 41:18-20 (December, 
1967). 


9.10 Hama, G: Calibration of Alnor Velometers, American 
Industrial Hygiene Association Journal (December 
1958). 


9.11 Hama, G.; Curley, L.S.: Instrumentation for the Meas- 
urement of Low Velocities with a Pitot Tube. Air 
Engineering (July 1967) and American Industrial Hy- 
giene Association Journal (May—June 1967). 


Chapter 10 


SPECIFIC OPERATIONS 


The following illustrations of hoods for specific operations order. 


are intended as guides for design purposes and apply to usual 
or typical operations. In most cases, they are taken from 
designs used in actual installations of successful local exhaust 
ventilation systems. All conditions of operation cannot be 
categorized, and because of special conditions (i.e., cross- 
drafts, motion, differences in temperature, or use of other 


means of contaminant suppression), modifications may be in 


Unless it is specifically stated, the design data are not to be 
applied indiscriminately to materials of high toxicity, e.g., 
beryllium and radioactive materials. Thus the designer may 
require higher or lower air flow rates or other modifications 
because of the peculiarities of the process in order to ade- 
quately control the air contaminant. 


10.15 Filling Operations 


Barrel Filling 


Weigh Hood Assembly — Dry Material 


Group | Operation Print No. Old No. Page 

{10.10 Cleanrooms L 10-6 
| ~~ 

| Cleanroom Ducted Module VS-10-01 10-8 

[_ Cleanroom Pressurized Plenum VS-10-02 ds 10-9 


VS-15-01 


VS-15-10 


Cleanroom Return Air Arrangements VS-10-03 10-10 | 


Bag Filling VS-15-02 VS-301 10-13 
Bag Tube Packer VS-15-03 VS-302 10-14 


40-11 


= Weigh Hood Details — Dry Material VS-15-11 | 10-16 
Toxic Material Bag Opening VS-15-20 | VS-1001 ies 10-17 
Shaft Seal Enclosure VS-15-21 VS-210 10-18 
[samping Box VS-15-30 VS-211 | 10-19 
10.20 Foundry Operations 10-20 
Foundry Shakeout — Enclosing VS-20-01 VS-110 
VS-112 10-21 
Foundry Shakeout — Side Draft VS-20-02 VS-110 
| VS-111 10-22 
near Shakeout i VS-20-03 VS-112 10-23 
Shell Core Making 4 VS-20-10 VS-114 10-24 
Core Making Machine — Small Rollover Type VS-20-11 VS-115 _| 10-25 | 
10.25 Gas Treatment 10-26 
| Fumigation Booth VS-25-01 VS-924 {0-27 
Fumigation Booth Notes | VS-25-02 VS-921.1 | 10-28 - 
Ethylene Oxide Sterilizers VS-25-10 10-29 | 
Ethylene Oxide Sterilizer Notes VS-25-11 ie 10-30 
*] Ethylene Oxide Sterilizer Hood Details VS-25-12 10-31 


10-2 


Industrial Ventilation 


Pe ee re he a tae => tld” saga [I~ widens alia OI 
VS-35-41 10-50 


10.40 Low Volume-High Velocity Exhaust 


Systems 


Extractor Head for Cone Wheels and Mounted VS-40-01 VS-801 10-52 


Protection Only) 
Specialized Laboratory Hood Designs 


Oven Exhaust 


Points 


VS-35-40 


Group Operation Print No. Old No. Page 
10.30 Kitchen Equipment 10-32 
| Dishwasher Ventilation VS-30-04 | VS-912 10-33 
pete Range Hoods | VS-30-10 VS-910 10-34 
L _| Kitchen Range Hood + VS-30-14 VS-911 | 10-35 
Charcoal Broiler & Barbeque Pit Ventilation VS-30-12 VS-913 10-36 
10.35 Laboratory Ventilation al 10-37 
“Prvpica Laboratory Hood T VS-35-01 VS-203 10-40 
2 [Genera Use Laboratory Hood Notes VS-35-02 VS-205 eee Se 10-41 
Perchloric Acid Hood Notes VS-35-03 VS-205.1 10-42 
Work Practices for Laboratory Hoods VS-35-04 VS-205.2 10-43 
Biological Safety Cabinet — Class II, Type A VS-35-10 10-44 
| Biological Safety Cabinet — Class Il, Type B VS-35-11 1 alley 0-45 
Dry Box or Glove Hood for High Toxicity & VS-35-20 VS-202 | 10-46 | 
Radioactive Materials 
Horizontal Laminar Flow Clean Bench (Product VS-35-30 VS-918.2 | 10-47 
- Protection Only} 
Vertical Laminar Flow Clean Bench (Product VS-35-31 VS-918.1 10-48 


VS-206 


{= 
Hood for Cup Type Surface Grinder and Wire VS-40-02 “tT Vveen | 1053 | 
Brushes 
[Pneumatic Chisel Sleeve VS-40-03 VS-803 L 10-54 
Extractor Head for Small Radial Grinders VS-40-04 VS-804 10-55 
Extractor Hood for Disc Sander VS-40-05 VS-805 10-56 
Extractor Tool for Vibratory Sander VS-40-06 VS-806 ) 10.57 | 
Typical System Low Volume-High Velocity VS-40-20 VS-807 10-58 
10.45 Machining ] | 10-59 
{Metal Cutting Bandsaw = VS-45-01 VS-418 10-60 
High Toxicity Materials Milling Machine Hood VS-45-02 VS-209 | 10-61 
| Metal Shears High Toxicity Materials VS-45-03 VS-208 10-62 
_ eon Heading Machine Ventilation VS-45-04 i VS-919 10-63 
. Lathe Hood VS-45-05 VS-207 10-64 
10.50 Material Transport 10-65 
Bucket Elevator Ventilation VS-50-01 VS-305 10-66 


Specific Operations 10-3 


_{Induction Melting Furnace— Tilting 


Group Operation Print No. Old No. Page 
Bin & Hopper Ventilation VS-50-10 VS-304 | _-10-67 
L [conveyor Belt Ventilation VS-50-20 VS-306 10-68 
| Toxic Material Belt Conveying Head Pulley | VS-50-21 VS-1002 10-69 
Toxic Material Conveyor Belt Loading VS-50-22 10-70 
| Rail Loading VS-50-30 10-71 
| Truck Loading | VS-50-31 10-72 
= 
10.55 Metal Melting Furnaces 10-73 
Melting Furnace Crucible, Non-Tilt VS-55-01 VS-103 10-74 
Melting Furnace, Tilting VS-55-02 VS-106 10-75 
~[Meting Furnace — Electric, Top Electrode | VS-55-03 VS-105 10-76 
a wens Furnace — Electric, Rocking VS-55-04 VS-104 10-77 | 
Melting Pot & Furnace | VS-55-05 VS-906 10-78 
enue Melting Furnace — High Toxicity Material a VS-55-06 | VS-201 zi 10-79 
VS-55-07 10-80 


10.60 Mixing 


Pouring Station 
Fixed Position Die Casting Hood 
Mobile Hood, Die Casting 


V S-55-10 
VS-55-20 
VS-55-21 


VS-109 10-81 
VS-804 10-82 
VS-905 10-83 


Mixer and Muller Hood VS-60-01 VS-107 10-85 
a cs 
Air Cooled Mixer and Muller VS-60-02 VS-108 10-86 
Banbury Mixer VS-60-10 VS-901 10-87 
Rubber Calendar Rolls VS-60-11 VS-902 10-88 
Rube | 
| Roller Mili Ventilation | VS-60-12 VS-902.1 10-89 
i + 
| 10.65 Movable Exhaust Hoods 10-90 
_|Moveable Exhaust Hoods [ VS-65-01 10-91 
eal Cutting and Finishing VS-65-02 VS-909 10-92 
i | Hawley Trav-L-Vent Perspective Layout VS-65-03 10-93 
a Oa 
j 10.70 Open Surface Tanks 10-94 
“T 
{Open Surface Tanks | VS-70-01 VS-503 10-104 
Open Surface Tanks VS-70-02 VS-503.1 10-105 
_|Push-Pull Hood Design Data for Widths Up to 10’ VS-70-10 VS-504 10-106 
Push-Pull Hood Design Data VS-70-11 VS-504.1 10-107 
\- 
| Push Nozzle Manifold Pressure VS-70-12 VS-504.2 10-108 
_|Solvent Degreasing Tanks VS-70-20 VS-501 10-109 
{Solvent Vapor Degreasing I VS-70-21 VS-501.1 10-110 
ST 
10.75 Painting Operations 10-111 
| 
Large Paint Booth VS-75-01 VS-603 10-112 


10-4 Industrial Ventilation 


oo: 


Group [Operation Print No. Old No. | Page | 
{Small Paint Booth | VS-75-02 VS-604 | 10-113 | 
Trailer Interior Spray Painting oe VS-75-03 VS-605 | 10-114 
| Large Drive-Through Spray Paint Booth VS-75-04 VS-606 10-115 
| _{Paint Booth Vehicle Spray VS-75-05 a VS-601 10-116 
_{Dip Tank VS-75-06 <a) VS-502 [104 1 7 | 
{Drying Oven Ventilation | VS-75-20 VS-602 H| 10-118 
| jPaint Mix Storage Room | VS-75-30 | p1oti9 | 
10.80 Mechanical Surface Cleaning and 10-120 
Finishing 
| [abrasive Blasting Room VS-80-01 ai VS-101 | 1. | 
[Abrasive Blasting Cabinet VS-80-02 VS-101.1 10-122 
ie {Tumbling Mills VS-80-03 im VS-113 | 10-123 | 
Grinding Wheel Hood — Surface Speeds Above VS-80-10 VS-411.1 10-124 
| _ {6500 sfpm I . = 
Grinding Wheel Hood — Surface Speeds Below VS-80-11 VS-411 10-125 
b= Hs sfpm I =| wl | 
Surface Grinder VS-80-12 VS-417 10-126 
Core Grinder VS-80-13 VS-102 10-127 
Vertical Spindle Disc Grinder VS-80-14 VS-410 10-128 
Horizontal Double-Spindle Disc Grinder VS-80-15 VS-408 10-129 
Swing Grinder VS-80-16 VS-414 10-130 
Abrasive Cut-Off Saw VS-80-17 VS-404 10-131 
Hand Grinding Bench . VS-80-18 VS-412 10-132 
| Portable Chipping and Grinding Table - | VS-80-19 VS-413 10-133 | 
| __|Manual Buffing and Polishing | VS-80-30 ie VS-406 10-134 | 
Buffing Lathe VS-80-31 VS-407 10-135 
| | Backstand Idler Polishing Machine VS-80-32 0 VS-402 ie 0-136 i 
_{Straight Line Automatic Buffing | VS-80-33 | VS-405 10-137 
L {Circular Automatic Buffing | VS-80-34 [_ VS-404 2 10-138 | 
Metal Polishing Beit | VS-80-35 ; VS-403 10-139 
10.85 Vehicle Ventilation i | | 10-140 
[Service Garage Ventilation — Overhead VS-85-01 _| VS-907 | 10-141 
| Service Garage Ventilation — Underfloor VS-85-02 2) VS-908 10-142 
Exhaust System Requirements for Typical Diesel VS-85-03 VS-908.2 10-143 
L |Engines Under Load | 
[ Ventilated Booth for Radiator Repair Soldering | VS-85-10 L | 10-144 
10.90 Welding and Cutting | e | 10-145 
Welding Ventilation Bench Hood VS-90-01 VS-416 10-146 
| | Welding Ventilation — Movable Exhaust Hoods VS-90-02 VS-416.1 10-147 


Specific Operations 10-5 
Group {Operation Print No. |__ Old No. Page 
iis Production Line Welding Booth VS-90-03 10-148 
Torch Cutting Ventilation VS-90-10 a VS-916 10-149 
Robotic Application VS-90-20 10-150 

1 
Metal Spraying VS-90-30 VS-415 10-151 
10.95 Woodworking 10-152 
Band Saw VS-95-01 VS-706 10-153 

Floor Table Saw _| VS-95-02 [ 10-154 | 
Radial Arm Saw VS-95-03 VS-709 10-155 
Swing Saw VS-95-04 VS-707 10-156 

ae - 

| Table Saw Guard Exhaust VS-95-05 10-157 
Single Drum Sander VS-95-10 VS-705 10-158 

| 
a = ple Drum Sander VS-95-11 24 VS-704 10-159 
Disc Sanders VS-95-12 VS-703 =| 10-160 


Optional Jet Stripper for Disk Sander 


VS-95-12a 


VS-702.1 


Horizontal Belt Sanders VS-95-13 VS-702 10-162 
Horizontal Belt Sander, Push-Pull System VS-95-14 10-163 


10-161 


Jointers VS-95-20 VS-701 10-164 
Exhaust Plenum Retrofit for Orbital Hand Sander VS-95-30 10-165 
Auxiliary Exhaust Retrofit for Air Powered Orbital VS-95-31 10-166 
Hand Sander 
10.99 Miscellaneous Operations 10-168 
Screens VS-99-01 VS-307 10-169 
= Table Slot VS-99-02 VS-505 10-170 
Canopy Hood | VS-$9-03 VS-903 10-171 
Indoor Pistol and Small Bore Rifle Range VS-99-04 VS-914 10-172 
Ventilation 
| Fluidized Beds VS-99-05 VS-915 10-173 
Outboard Motor Test VS-99-06 VS-820 10-174 
Mortuary Table VS-99-07 10-175 
a 7 
Furniture Stripping Tank VS-99-08 10-176 


10-6 Industrial Ventilation 


10.10 CLEANROOMS 


U.S. Federal Standard 209E“°!° establishes standard 
classes of air cleanliness for airborne particulate levels in 
cleanrooms and clean zones. This standard is issued by the 
General Services Administration of the United States. While 
nominally a publication for use by federal agencies, FED- 
STD-209E has been adopted by American industry. It pre- 
scribes methods for class verification and monitoring of air 
cleanliness. It also addresses certain other factors that affect 
control of airborne contaminants. 


FED-STD-209E does not address the physical, chemical, 
radiological, or viable nature of airborne contaminants. It also 
does not address the occupational health concerns of employ- 
ees working in cleanroom environments. 


A cleanroom class is the statistically allowable number of 
particles, greater than or equal to 0.5 micrometers in size, per 
cubic foot of air. Cleanroom classes are shown in Table 
10.10.1. 


In order to meet the class limits, a high efficiency particu- 
late air (HEPA) or ultra low penetration air (ULPA) filter is 
required. A HEPA filter is a disposable, extended-media, 
dry-type filter in a rigid frame with a minimum particle 
collecting efficiency of 99.97% for 0.3 micrometer, thermally 
generated dioctylphthlate (DOP), or specified alternate, aero- 
sol particles at a maximum clean resistance of 1.0 "wg when 
tested at rated air flow capacity. An ULPA filter is a dispos- 
able, extended-media, dry-type filter in a rigid frame with a 
minimum particle collecting efficiency of 99.999% for par- 
ticulate diameters between 0.1 and 0.2 micrometers in size. 


Military specifications(°!°% and publications®!° by the 
Institute of Environmental Sciences (IES) define HEPA and 
ULPA filter construction. Filters having an efficiency even 
higher than an ULPA filter are available from some compa- 
nies specializing in cleanrooms and air filtration. 


TABLE 10.10.1. Class limits in particles per cubic foot of size 
equal to or greater than particle sizes shown.* 


Room Measured Particle Size (Micrometers) 
ees 0.1 0.2 03 05 5.0 
1 35 is: 3 1 NA 
10 350 7 30 10 NA 
100 NA* 750 300 100 NA 
1000 NA NA NA 1000 7 
10,000 NA NA NA 10,000 70 
100,000 NA NA NA 100,000 700 


*The class limit particle concentrations shown are defined for class purposes only 
and do not necessarily represent the size distribution to be found in any particular 
situation. 


*Not applicabie. 


The primary design considerations for cleanrooms are the 
supply air flow rate, the air flow patterns within the clean- 
room, the method for recirculating the air from the cleanroom 
and the filter efficiency. 


Air is supplied to the cleanroom by an air handling system 
containing the components needed for heating, cooling, and 
humidity control. Noise is readily transmitted to the clean- 
room so very slow fan speeds, vibration isolation and noise 
control devices are important design considerations. The air 
circulation system will also contain two or three stages of 
prefiltration. This allows the final filters in the cleanroom 
ceiling to remain in place for very long periods of time. A final 
filter life of ten years or more is typical for Class 100 and 
better cleanrooms. 


Air from the supply system enters the cleanroom through 
either a ducted module or a pressurized plenum. VS-10-01 
shows the ducted module arrangement. Ducted modules con- 
taining HEPA or ULPA filters are connected to the main air 
supply duct by flexible branch ducts. The modules usually 
contain an internal baffle for balancing the air exhaust which 
must be at a uniform velocity across the face of the filter. The 
ducted modules are mounted in a T-bar grid and sealed with 
either solid gaskets or a liquid gel sealant. The ducted mod- 
ules, because of long filter life, usually are considered to be 
throwaway items; however, some arrangements do permit the 
replacement of filters from within the cleanroom. A ducted 
module system offers maximum flexibility for cleanroom 
modification. 


VS-10-02 shows a pressurized plenum arrangement. A 
heavy-duty grid system is suspended from the ceiling with 
suspension rods and the HEPA or ULPA filters sealed in the 
grid with liquid gel or solid gaskets. The entire plenum is 
pressurized by the air supply system to allow a uniform flow 
of air through the filters to the cleanroom below. A pressur- 
ized plenum system will usually cost less than a ducted 
module system for large cleanrooms. 


VS-10-03 shows raised floor and low sidewall arrange- 
ments. Air is returned through a utility chase to the cleanroom 
supply air system. To provide better particulate control, the 
raised floor arrangement is preferred. The low sidewall return 
should not be used for vertical downflow cleanrooms more 
than 14 feet wide in order not to disrupt laminar flow at the 
work area. 


IES-RP-CC-006-84"°'°4) contains testing methods for 
characterizing the performance of cleanrooms. It defines 
terms having special meaning and describes test procedures 
to assure proper cleanroom operation. Uniform air flow is 
defined as unidirectional with all velocity readings within 
20% of the average velocity of the work area. The air velocity 
at the work area is generally about 100 fpm; however, design 
conditions may require velocities of 10 fpm or lower. 


REFERENCES 


10.10.1 U.S. General Services Adminstration: FED-STD- 
209E, Federal Standard, Cleanroom and Work Sta- 
tion Requirements, Controlled Environment. 
Federal Supply Service, General Services Admini- 
stration, Washington, DC (June 15, 1988). 


10.10.2 U.S. Army: MIL-F-51068(EA), Specification Fil- 
ters, Particulate, High-Efficiency, Fire Resistant, 


10.10.3 


10.10.4 


Specific Operations 10-7 


Biological Use, General Specifications For. Com- 
mander, U.S. Army Armament Research and Devel- 
opment Command, Aberdeen Proving Ground, MD 
(October 4, 1982). 


Institute of Environmental Sciences: JES-RP-CC- 
001.3, HEPA and ULPA Filters. IES, Mount Pros- 
pect, IL. 


Institute of Environmental Sciences: IES-RP-CC- 
006, Testing Cleanrooms. IES, Mount Prospect, IL. 


10-8 Industrial Ventilation 


Air is supplied to the ducted modules from the air distribution duct 

through flexible branch ducts, which are secured at the bottom ends 
by clamps. A doamper (not shown) on the inside of the collar allows 
balancing of the air flowing from the module to the cleanroom. 


Ducted modules are mounted in 2’ x 4’ grids and sealed with gaskets 


or a liquid gel sealant. Tear drop or recessed lighting (shown) 
provides illumination. 


J AMERICAN CONFERENCE | CLEANROOM 
| OF GOVERNMENTAL — | DUCTED MODULE 
J INDUSTRIAL HYGIENISTS BRE—7B—gG FEES OT OT 


Specific Operations 10-9 


-—-H E PLA. Filter E 
(shown removed : 


Distribution plate— 
Supply air 


Plenum space- Lier <_ 


Suspended fra me/ 
system 


~~ Light fixture 


‘| DR Vertical airflow ——~¢ly 
through room 


Supply air enters a pressurized plenum and strikes a distribution plate. Each 
2' x 4’ opening in the support structure contains a HEPA filter. Filters are 
sealed around the perimeter with gaskets or a fluidic sealant. The framing 
structure is supported from the plenum ceiling by suspension rods. Tear drop 
lighting is shown. 


CLEANROOM 
PRESSURIZED PLENUM 
Pee 90 Pee VS 10-02 | 


NFERENCE 
OF GOVERNMENTAL 


I INDUSTRIAL HYGIENISTS 


AMERICAN CO 


10-10 Industrial Ventilation 


Raised floor 


Utility compartment — 


Raised floor with depressed slab. Air is returned through a_ utility 
compartment to the air supply system. 


Utility compartment —_ 


oe 


Low sidewall grille return through a utility compartment to the 
air supply system. Room width is limited to 14 feet if laminar 
air flow is to be achieved. The distance from the top of the 
grille to the floor should not exceed 18 inches. 


AMERICAN CONFERENCE CLEANROOM RETURN 
} OF GOVERNMENTAL | AIR ARRANGEMENTS 
J INDUSTRIAL HYGIENISTS EE“ 7s=97—— Rane Te 


10.15 FILLING OPERATIONS 


Filling operations have special considerations that should 
be addressed when designing hoods. An enclosed space is not 
empty but rather is filled with air. When material enters the 
space, it forces the air out which in turn can carry some of the 
material with it. Also, additional air can be entrained by the 
material stream entering the enclosed space. This effect is a 
function of the size of the particles and the distance the 
material must fall. These two effects must be considered when 
designing hoods for material handling situations. 


If there are any openings in the walls of the container which 
is being filled, some "splashing" of the material can occur. 
This can lead to loss of material through cracks and openings 
in the receiving vessel. The design of the ventilation system 
should take this effect into account. 


The proper choice of exhaust flow rate is critical. Iftoo little 
air is exhausted, the air displaced by the falling material may 
exceed the exhaust rate and the contaminant may not be 
adequately controlled. If too much air is exhausted, excess 
material could be entrained into the exhaust air stream. As this 
material often is the product, excess product loss could occur. 


VS-15-01 illustrates four different ways of controlling 
barrel or drum filling operations. VS-15-02 illustrates bag 
filling and weighing. VS-15-03 depicts a bag tube packer. 
VS-15-10 and VS-15-11 depict a weighing hood where dry 
materials are removed from a bulk pack and weighed into 
smaller bags.('?'53) These smaller bags are then packed into a 


Specific Operations 10-11 


container. Bags containing toxic materials can be opened 
within an enclosing hood such as shown on VS-15-20.0°'5.9 


VS-15-30 shows how to extract a toxic liquid from a 
process line or vessel for analysis and VS-15-21 shows a 
possibility of controlling leaks around rotating shafts that 
enter containers.“ 


REFERENCES 


10.15.1 Hama, G.M.: Ventilation Control of Dust from 
Bagging Operations. Heating and Ventilating, p. 91 
(April, 1948). 


10.15.2 Cooper, T.C.: Control Technology for a Dry Chemi- 
cal Bagging and Filling Operations. Monsanto Ag- 
ricultural Products Co., Cincinnati, OH (1983). 


10.15.3  Gressel, M.G.; Fischback, T.J.: Workstation Design 
Improvements for the Reduction of Dust Exposures 
During Weighing of Chemical Powders. Applied 
Industrial Hygiene 4:227—233 (1989). 


10.15.4 Goldfield, J.; Brandt, F.E.: Dust Control Techniques 
in the Asbestos Industry. A paper presented at the 
American Industrial Hygiene Conference, Miami 
Beach, FL (May 12~17, 1974). 


10.15.5 Langner, R.R.: How to Control Carcinogens in 
Chemical Production. Occupational Health and 
Safety (March—April, 1977). 


10-12 Industrial Ventilation 


Close clearance 


100 cfm /tt* barrel top (minimum) Q = 150 cfm /ft” of open face area 
Minimum duct velocity = 3500 fpm Minimum duct velocity = 3500 fpm 
= 1.78 VP, + 0.25 VP, h, = 0.25 VP, (45° toper) 


e 


Feed spout —~ 
4° min. dia. 


Flex duct--— 


50 cfrn x drum diam. (ft) = 300-400 cfm 


Minimum duct velocity = 3500 fpm Minimum duct velocity = 3500 fom 
he = 0.25 VPy he = 0.25 VPy 


€ 


Note 1: Air displaced by material feed rate may require higher exhaust flow rates. 


Note 2: Excessive air flow can cause loss of product. 

AMERICAN CONFERENCE pies ae 

OF GOVERNMENTAL : BARREL FILLING 
INDUSTRIAL HYGIENISTS 


Specific Operations 10-13 


Q = 400-500 cfm —- non-toxic dust 
1000-1500 cfm — toxic dust 


Minimum duct velocity = 3500 fpm 
he = 0.25 VPy 


Care must be taken such that too much air 
is not used, as valuable product will be 
pulled into the exhaust system. 


Reference: 10.15.2 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
fF INDUSTRIAL HYGIENISTS 


BAG FILLING 


“TFIGURE. 


10-14 Industrial Ventilation 


Ss a oa , 
SOK 
Spill hopper ) 3 pare oy OF 


—} 


Q = 500 cfm per filling tube 
S00 cfm at Feed hopper 
950 cfm at Spill hopper 


Hol 


Minimum duct velocity = 3500 fpm 


he = 0.25 VPq for take-off at A and C 
1.0 VPy for take-off at B 


: Reference: 10.15.2 


| AMERICAN CONFERENCE | 
| OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


BAG TUBE PACKER | 
da a LSE 


Specific Operations 10-15 


f Air Shower 


/~— Dry Material 
Container Hood 


Dry Material Container 


— Scale 


Bag Container 


See VS~-15~-11 for design details 


erence 10.15.3 


WEIGH HOOD ASSEMBLY 
: DRY MATERIAL : 


MERICAN CONFERENCE J 
OF GOVERNMENTAL 
fF INDUSTRIAL HYGIENISTS 


10-16 Industrial Ventilation 


Q,g = 50 cfm/tt* of face open area. 


L and W to fit operation 
Minimum duct velocity = 3500 fpm 
he= 1.78 VR, + 0.25 VPy 


rp Dry material container hood is extension of 
booth slot 


100 Ly cfm 


Ls = 3 feet. (Can be longer if required to fit 
workstation but do not exceed 1/2 booth 


length; L) 


0.25" pegboard or equivalent, 20 percent 
maximum open area. 


DRY MATERIAL CONTAINER HOOD 
Hood is extension of booth slot. An additional 
takeoff(s) may be used if required for hood air flow 
distribution. 


Airflow and hood slot design per VS-15-01 


—12"” to 24° Diameter 


TS o> 
“~ 24 Maximum 


peeelce 10. 19. » 


AMERICAN ‘CONFERENCE WEIG H HOOD DETAILS 
OF GOVERNMENTAL : DR r eae TAL 


INDUSTRIAL HYGIENISTS [RE a [Le “VS “15—11 


Specific Operations 


To exhaust 


To exhaust 


Compactor to take —— Light inside hood 
fiber bags ——--———-~___ 

Open grille work shelf - oer . ; 
(under hood) “Ie po 6 Rubber Curtain 


lL Hopper connected to 


Q = minimum 250 cfm/ft? of open area 
Minimum duct velocity = 3500 fpm 
he = 0.25 VP, 


Reference: 10.15.4 


AMERICAN CONFERENCE / me ee L 
J OF GOVERNMENTAL OE Sen 
| INDUSTRIAL HYGIENISTS k=— ES TISL OT 


screw feed, chute, etc. 


10-17 


10-18 Industrial Ventilation 


~ \/_ [ieee 


Process 
fluid 


4. Optional drain plug 


Optional slinger g 
Used to prevent the process fluid 
from creeping along the shaft. 


Q = 500 cim/it’ of open area 
(typically 10-40 cfm) 
Note: Sufficient air must be provided to 

dilute flammable gases and/or vapors 
to below 25% of LEL. See Chapter 2. 
Duct velocity = 2000 fpm 
he = 1.78VP, + O.25VP,y 


Note: Similar hood is appropiate for unions 


40.15.5 


Reference: 
AMERICAN CONFERENCE 
OF GOVERNMENTAL 

FS INDUSTRIAL HYGIENISTS 


SHAFT SEAL ENCLOSURE 


Specific Operations 10-19 


_-—— Process line or vessel 


sampling valyv 


-—~ Slots or perforated plate. 


Door ——/ 
Swing out or vertical sliding 
Interlock desirable to prevent 


sample extraction unless door 
is closed. 


Q = 125 cfm/ft* of open area (door area) minimum 
Duct velocity = 2000 fpm 
h, = 1.78 VES + 0.50 VP, 


e 


NOTE: Sufficient air must be provided when door closed 
to dilute flammable gases and or vapors to 25 % 
of LEL. See Chapter 2 


Reference: 10.15.5 


AMERICAN CONFERENCE 
| OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


SAMPLING BOX 


10-20 Industrial Ventilation 


10.20 FOUNDRY OPERATIONS 


Foundry operations include many operations common to 
other industries. Some of these operations are covered in the 
following subsections of this chapter: 


10.45 Machining 

10.50 Material Transport 
10.55 Metal Melting 
10.60 Mixing 

10.80 Surface Cleaning 
10.90 Welding and Cutting 


This subsection addresses operations that are more unique 
to the foundry industry: casting shakeout and core making. 


10.20.1 Casting Shakeout: Foundry shakeout ventilation 
rates depend on the type of enclosure and the temperature of 
the sand and castings. The enclosing shakeout hood (VS-20- 
01) requires the smallest air flow rate. The side draft shakeout 
hood (VS-20-02) requires additional air flow rates but pro- 
vides improved access for casting and sand delivery and for 
casting removal. The downdraft shakeout (VS-20-03) is the 
least effective in controlling contaminant and requires the 


highest ventilation rates. It is not recommended for hot cast- 
ings. The shakeout hopper below the shakeout table requires 
additional exhaust ventilation equivalent to 10% of the 
shakeout hood exhaust rate. 


Particular attention should be paid to the conveyor remov- 
ing sand from the shakeout. This conveyor requires hoods and 
ventilation as described in Section 10.50. 


Rotary tumble mills used for shakeout should be treated as 
an enclosing hood with a minimum inward velocity of 150 
fpm through any opening. 


10.20.2 Core Making: Core making machines require 
ventilation to control reactive vapors and gases such as amines 
and isocynates that are used in the core making process. A 
minimum capture velocity of 75 fpm is required. However, a 
ventilation rate as high as 250 cfm/ft? of opening may be 
necessary for adequate control of contaminant emissions. 
When cores are cured in ovens, adequate ventilation control 
of the oven is required. 


REFERENCES 


10.20.1 American Foundrymen's Society, Inc.: Foundry 
Ventilation Manual. AFS, Des Plaines, IL (1985). 


ENCLOSING HOOD 


Provides best conirol with least flow rate 
Minimum duct velocity = 4000 fpm 
he= 0.25 VPy 


<~ Working openings, 


Specific Operations 10-21 


- 
BS keep as small as 
wr, 
| 


(a) 


Mold 
conveyor. 


ct — | | a me 


— Castings 
out here 


~ Shakeout 


_ Shakeout exhaust, minimum * 


: Type of hood Hot castings 


- Co ool castin 19s = 


(Enclosing. ** VS=20-04 200 cfm /ft” opening 


At least 200 cfm/ft* 
alg area 


200 Scot opening 


At least 150 cfm/ft* 
grate area 


Two sides and 1/3 456 


area enclosed** VS-20-02_ 300 cfm/ft grate oreo 


275 cfm /ft? grate area 


Side hood (as shown or 400-500 cfm/ft’ grate 
equivatent) ae Vee 20- “02 area 


= dL sinnaistn 


2 
350-400 cfm/ft grate 
area 


Double side hood** 400 cfm/ft* grate area 


t i a 


VS-20-02 


300 stm/ft? ele area 


*Choose higher values when 
(1) Castings are quite hot 
(2) Sand to metal ratio is 
(3) Cross—¢drafts are high 


iow 
** Shakeout hoppers require an additional 10% exhaust. 
AMERICAN CONFERENCE 


OF GOVERNMENTAL 
INDUSTR ale Ne 


FOUNDRY SHAKEOUT 
ENCLOSING 


FIGURE 


10-22 Industrial Ventilation 


<—-—- Moveable panels to secure 


desired distribution. 
cere L Channel iron guard Optional tes 
[ 


Neg Bie 
ok : fake Sot. 
THT 0.2 Wye | \ 

H=L | 

: a 

{ dissent 


Baffle to edge — Res Rigidly 

adi RT | ] 1of grate "oS = braced 

moe ; 4 

Shakeout 

a L _ pao Wea | 
. \ 


Minimum practical clearance _\ 


L-—- Velocity through openings 2000 fom 


SIDE-DRAFT HOOD 


Minimum duct velocity = 4000 fpm. 
he = 1.78 VR + 0.25 VP, 


<-~ Blank wall in this position is 
almost as good as double hood. 


ASK 
Minimize 4 


clearance 4+} 
~ Rigidly braced 


} 


DOUBLE SIDE-—DRAFT 


Proportions same as single side-draft hood except for overhang. 


Minimum duct velocity = 4000 fpm 
Slots sized for 2000 fpm 
he = 1.78 VP, + 0.25 VP, 


: See VS-20-01 for exhaust rates 


FOUNDRY SHAKEOUT 
SIDE DRAFT 


DATE 10-90 fricuRE VS— 20-02 


| AMERICAN CONFERENCE 
f OF GOVERNMENTAL 
s INDUSTRIAL HYGIENISTS 


Specific Operations 10-23 


If feeder enclosure is over 10 feet long, provide exhaust 
at hopper. See VS-50-10 and VS-99-01 — 


we = es x 
Hie e karte Loe, —_Enclose pan feeder or belt completely. —— 


4 x duct area 


mud 


Exhaust at transfer to elevator. 


SIDE VIEW 


END VIEW 


ELEVATOR 
See VS-50-01 


HOPPER EXHAUST DETAIL 


FOUNDRY SHAKEOUT 


| AMERICAN CONFERENCE 
| OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


a a 


10-24 Industrial Ventilation 


PLES ERM LH LY Lb) 
PIT TTA Eyed yy : 
HHH Pp dtp epi ily 2L 
TEE US PUTER eyp Up ey Ely | — oo 
VUE PEEP U Tp UPL epee ly | 


\— Slotted 
side draft 
hood. -—— 


i 


fae oO Sulls 
operation 


Use side baffle on canopy hood as 


Canopy hood: Q = 250 cfm/ft* canopy — single unit 


150 cfm /ft* canopy — double unit 
h, = 0.25 VP, 


e 


Note: Slotted side draft hoods required to remove 
smoke as hot cores emerge from machine. 
Minimum capture velocity = 150 fpm 


Side draft hood: Q = 150(10X* + A) where A equals hood area 
he = 1.78 VP; + 0.25 VPy 


Note: Conveyor or cooling area require ventilation for 
large cores. Scrap conveyor or tote boxes may also 

require additional ventilation. 

Minimum duct velocity = 3500 fpm 


AMERICAN CONFERENCE 
| OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


SHELL CORE MAKING 


SSaEmNTIaT Tian Ueeaeen ecomearrameTte 


Specific Operations 10-25 


| Top view of take-off 
; connection 


Roll-over handle = 


Hood. Closed on ends, 
top and sides. 7 


- Rotating 
connection 


Seal around shaft =) 


SIDE VIEW 


2 
Q = 200 cfm/ft of open face area 
Minimum duct velocity = 3500 fpm 
he = 0.25 VP, 


Elbow and rotating connection losses 
not included. 


1 AMERICAN CONFERENCE CORE MAKING MACHINE 
| OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


10-26 Industrial Ventilation 


10.25 GAS TREATMENT 


The handling of gas cylinders for industrial operations 
requires special attention. In addition to the potential safety 
problems associated with transportation and use of com- 
pressed gas cylinders, the gas inside the cylinders can escape 
through leaky valves and fittings. During connection and 
disconnection of the gas lines, due to the operating pressures, 
gas can be released. 


This section of VS-prints illustrates uses of toxic gases 


during fumigation (VS-25-01] and -02) and during ethylene 
oxide sterilization (VS-25-10, -11 and -12). 


REFERENCES 


10.25.1 Mortimer, V.D.; Kercher, S.L.; O'Brien, D.M.: Ef- 
fective Controls for Ethylene Oxide — A Case 
Study. Applied Industrial Hygiene 1(1):15-20 
(1986). 

10.25.2 Hama, G.M.: Ventilation for Fumigation Booths. 
Air Engineering (December 1964). 


Specific Operations 10-27 


A Air 
inlet door + 


Ventilation Rates 


Allow 60 minute purge time 
Ventilation rate must be 20 air changes per hour or greater. 
Design must provide: 


(a) 500 fpm velocity or greater through air inlet door 
when large access door is closed 
and 


(b) at least 100 fpm through all openings when large 
access door is open. 


: Reference: 10.25.2 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


FUMIGATION BOOTH : 


10-28 Industrial Ventilation 


NOTES: 


1. Provide an air inlet with automatic damper closure; damper must be interlocked with 
fan circuit to open only when fan is turned on. Size opening for a minimum velocity of 
500 fpm. Air inlet must be located so purge air sweeps entire booth. 


. Loading door must be opened only when booth has been completely purged. Provide 
gaskets, screw clamps, and brackets for applying uniform pressure for a gas-tight fit. 


. Provide ventilated cabinet for gas cylinders in use and being stored. Fan must be on 
continuously and exhaust approximately 500 cfm to produce a negative pressure in 
the cabinet when the doors are closed. 


. Provide nozzle openings for introducing fumigant gas. A circulating cabinet fan should 
also be provided for obtaining good mixture of fumigant gas. 


. Mechanical fan damper must be provided that closes tightly when fan is shut off 
during fumigation and opens when fan is turned on. Damper controls should be 
interlocked with fan controls. 


. Fan for ventilating fumigation booth must be sized to dilute air to safe limit in required 
time. Use vertical, outside, discharge stack away from windows, doors, and air 
intakes. 


. Fumigant gas cylinder cabinet fan must run continuously. 


. Control switches for fan and lights and an air flow switch-actuated pilot light are 
recommended. 


. Red warning light to indicate booth is under fumigation as a protection against 
careless entry is recommended. 


. To facilitate penetration of fumigant gas and subsequent airing out, mattresses should 
be loaded with separators to allow free air space around each mattress. 


. Fumigants with no odor-warning properties should be used together with an odor- 
indicating chemical. 


. Where toxic fumigants are used, a leak test should be made on the booth. The booth 
first should be tested by lighting several large smoke candles in it with doors and 
dampers closed. Leaks can be noted by the presence of smoke at the point of escape. 
Where highly diffusible toxic gases are used, an additional test should be made with 
the booth under charge, at doors and dampers, with a sensitive detecting meter or 
sampling device. 


AMERICAN CONFERENCE 
f OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


: FUMIGATION BOOTH NOTES 


Specific Operations 10-29 


Q.(Note 1) Qpy(Note 2) Q,(Note 5) 
7° | Qe(Note 6) 
| 


/ 


l Qe(Note 6) 
Qy(Note SA 


7 


Lipp St Gi at ~ Sterilizer 
Anti Sipnon PE ORES, . Bags, ob ds 
2 + GC see Vos 20-1) 
Ethylene oxide — Air Gap ( ) 
cylinders 


eee . EXHAUST FLOW RATE 
rYPE OF EQUIPMENT (Q,) CEM 


Electrically heated | See VS—25-11 for notes. 
gos sterilizers (< 10 ft? ) 0.10 


af See VS—25-12 for notes. 
Stearn heated gas i 
or steam sterilizers 


Aeratoe and instrument 
wasner units 


Reference: LO Zou 


AMERICAN CONFERENCE Stes acne, Peimeuas ecient 
| or GOVERNMENTAL | ETHYLENE OXIDE STERILIZERS 
| INDUSTRIAL HYGIENISTS 


10-30 Industrial Ventilation 


NOTES: 


1. The ethylene oxide (EtO) supply cylinders should be placed in a ventilated cabinet ora 
partially enclosed hood with an exhaust rate, Qe, of at least 100 cfm/ft2 of open area. 


. The anti-syphon air gap in the sterilizer evacuation drain line should be enclosed and 
ventilated. The enclosure should have one or two openings to allow air, Qp, to enter 
and to prevent liquid, which might back up from the drain, from reaching the sterilizer 
evacuation line. In lieu of a greater value specified by the sterilizer/vacuum pump 
manufacturer, Qy should be approximately 50 cfm and the openings sized to maintain 
approximately a 600 fpm face velocity. 


. The overpressure relief valve should be vented to carry EtO out of the building if it 
should ever open. With a sealed line connecting the valve with the ventilation duct, 
there will be no ventilation volume, Q,, except when the valve opens. Consult the 
sterilizer manufacturer for the proper size of this line; too much resistance could 
interfere with proper venting of the chamber. 


. A hood should be placed above the sterilizer door to remove EtO rising from the 
chamber when the sterilizer door is “cracked” open a few inches for approximately 15 
minutes before the sterilized items are removed from the chamber. See VS-25-12 fora 
discussion of the exhaust volume, Q,, requirements. 


. If an aerator is installed, its door should be hinged, and it should be placed beside the 
sterilizer so that the doors of the gas sterilizer and aerator open away from each other 
to facilitate transferring the sterilized items. Consult the manufacturer for the required 
air flow, Qa. 


. The room behind the wall enclosing the sterilizer(s) and other equipment should be 
exhausted adequately to handle the air driven to the ceiling by the thermal gradients 
caused by the heated equipment. The ideal arrangement would be to have a properly 
sized vent above each piece of heated equipment. The total Qg should be the sum of 
the values for each piece of heated equipment (see VS-25-10) plus 100 cfm/ft2 of open 
area for transfer vents placed in the upper portion of the room. However, federal 
hospital standards specify that, for a recess room containing a gas sterilizer, the 
votume exhausted in one hour should be at least ten times the room volume. Transfer 
vents placed in the lower portion of the room will help the influx of air to supply the 
thermal air currents and would not add to the total exhaust requirement. 


. All air that could contain EtO should be exhausted through a ventilation system which 
does not have vents in any other rooms. The discharge of the fan on the roof should be 
located so that the exhausted air will not re-enter the building or expose people 
outside the building. This ventilation system should have a flow sensor/alarm to warn if 
it is not functioning properly. If there is the possibility of lint in the exhausted air, use 
a differential pressure sensor or some other type that will not be clogged or stuck open 
by the accumulation of lint. 


AMERICAN CONFERENCE | eTHYLENE OXIDE STERILIZER 
OF GOVERNMENTAL — | 
INDUSTRIAL HYGIENISTS 


BROT 


Specific Operations 10-31 


slot face to be at 45°-90° 
angle with plane of enclosure § 


Q = 75 cfm/ft slot length 
slot 


= 2” for airing out sterilizer 
chamber. 


= 100 LLood ™ 


= 12” to 24 


Hood 


3 


Lod X + 0.6641 Oo og 


Y = 2” for airing out sterilizer 
chamber. 


CANOPY HOOD 


| AMERICAN CONFERENCE | eT HYLENE OXIDE STERILIZER | 
| OF GOVERNMENTAL HOOD DETAILS 


| INDUSTRIAL HYGIENISTS EST ROS TST a 


10-32 Industrial Ventilation 


10.30 KITCHEN EQUIPMENT 


The purpose of an exhaust system for kitchen equipment is 
to control] heat, humidity and grease vapor released into the 
space by cooking or dishwashing equipment. A secondary 
consideration is the control of combustion products associated 
with the heat source which may be vented separately or 
through the hood itself. 


National Fire Protection Association (NFPA) Standard 
96030.) describes grease filter construction as well as hood 
construction necessary to maintain hood integrity in the event 
of a fire. Welded seam construction is preferred and some- 
times required by public health authorities to assure cleanabil- 
ity and ease of maintenance. The National Sanitation 
Foundation Standard No. 4°#°2 also lists hood construction 
requirements for cleanability and integrity in the cooking and 
food zones within the hood. In all cases, the local health 
authorities having jurisdiction should be consulted for con- 
struction requirements prior to hood fabrication. 


Fire is a primary concern with all cooking equipment. Each 
hood will require some type of fire suppression consistent 


with local fire code requirements. The system selected must 
not Compromise sanitation or endanger workers due to loca- 
tion or system activation. Hood or duct penetrations by fire 
suppression piping, etc., must be sealed to prevent short 
circuiting of air or loss of fire arrestance. 


For high temperatures situations such as exposed flames or 
charcoal, the grease filters must be sufficiently removed from 
the heat source to prevent ignition. Fan selection may require 
use of high temperature fan components and consideration of 
the effect of change in air density. 


REFERENCES 


10.30.1 National! Fire Protection Association: Standard for 
the Installation of Equipment for the Removal of 
Smoke and Grease-Laden Vapors from Commercial 
Cooking Equipment, (Standard 96). NFPA, Quincy, 
MA (1987). 


10.30.2 National Sanitation Foundation: Commercial Cook- 
ing and Hot Food Storage Equipment, Standard No. 
4. NSF, Ann Arbor, MI (1986). 


Specific Operations 10-33 


Pitch duct 
toward hood 


Dishwasher 


Q = 250 cfm/ft® of door area— each end 
Minimum duct velocity = 1000 — 3000 fpm 
he = 0.25 VP, 


CANOPY HOODS 


2” slot 
around hood 


—e-| haa 6” 
Dishwasher 


6" 


Q = 150 cfm/ft* of door area (150WH) 
each end 
Minimum duct velocity = 1000 — 3000 fpm 
SLOT HOODS OH, =: 100: VPI 0.25 VP, 


Dishwasher 


Q = 150 cim/ft 2 of entrance and exit area 
Minimum duct velocity = 1000 ~ 3000 fpm 


= =) 
EXHAUSTED VESTIBULES he 0.50 VP, 


Note: If direct exhaust connections are provided from dishwasher body, cap these 
connections and use external hoods. 


AMERICAN CONFERENCE 


p OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


DISHWASHER VENTILATION 


SO 


10-34 Industrial Ventilation 


Ducts 6’ on center (max.) 


rm for large hoods 
| a SS | Grease fi 


[Removable | 

eB pane drip pan 
+. overhang on RO W—— 
three sides f 


Cooking equipment 
| aes ieee! 
HOOD AGAINST WALL 
Q = 80 cfm/ft’ of hood area (80 WL) 
Not less than 50 cfm/ft* of face area (50 PH) 
P = perimeter of hood = 2W + L 
Duct velocity = 1000 — 4000 fpm, to suit conditions 
h, = (filter resistance + 0.1") + 0.50 VP,( straight take off) 


e 


h, = (filter resistance + 0.1") + 0.25 VP,{ tapered take off) 


he 6’ maximum 4 
; | Grease filters— 


A5°-60° 


eS 


6" min. 
-overhang on 
all sides 


Island cooking area 


a 


ISLAND TYPE HOOD perimeter of hood | 
2W + 2L 


125 cfm/ft* of hood area (125 WL) 

Not less than 50 cfm/ft’ of face area (50 PH) 
Minimum duct velocity = 1000 — 4000 fpm, to suit conditions 
h, = (filter resistance + 0.1") + 0.50 VP, (straight take off) 
h, = (filter resistance + 0.1") + 0.25 VP, (tapered take off) 


e 
e 


ote: See VS~30-11 for information about filters and fans for range hoods. 
AMIEERICAN CONFERENCE 
OF GOVERNMENTAL 


KITCHEN RANGE HOODS 


Specific Operations 10-35 


Ducts 6’ on center maximum Plenum — 


Filters 
= 20° minimum 4 


Face or ends can be 
opened for filter 


removal > eS “< 
d 
1? max oo) 


sei—back 


Filter mounting height 
See note 4 below 


yl 
\} 


i Closed ae a 


Cooking equipment desirable | 


LOW SIDE WALL HOOD 


Q = 200 cfm/lineal ft of cooking surface (200L) 

Minimum duct velocity = 1000 ~ 4000 fpm, to suit conditions 
= (filter resistance + 0.1”) + 0.50 VP, (straight take off) 

= (filter resistance + 0.1") + 0.25 VP, (tapered take off) 


NOTES FOR KITCHEN HOODS 


Filters: 1. Select practical filter size. 
Determine number of filters required from manufacturer's data. 
(Usually: 2 cfm maximum exhaust for each square inch of filter area.) 
install at 45°-60° to horizontal. Never horizontal. 
Filter mounting height (Reference 10.30.1) 
a. No exposed cooking flame —-~- 18 minimum to lowest edge of filter. 
b. Charcoal and similar fires ~-~- 4’ minimum to lowest edge of filter. 
Shield filters from direct radiant heat. 
Provide removable grease drip pan. 
Clean pan and filters regularly. 


Use upblast discharge fan. Downblast is not recommended. 
Select fan for design Q and SP resistance of filters and duct. 
Adjust fan specification for expected exhaust air temperature. 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


KITCHEN RANGE HOOD 


10-36 Industrial Ventilation 


e , = Grease filters 


-—~Metal sides 


rikewer edge of filters 
at least 4’-6” above fire 


CHARCOAL BROILER 


Q = 100 LH 
Minimum duct velocity = 1000 — 3000 fpm 

he (filter resistance + 0.1") + 0.50 VP (straight toke off) 
he = (filter resistance + 0.17) + 0.25 VP (tapered take off) 


iT 


# Glass to be 
| pyrex or high 


—Grease filters 


-——Metal sides 


Lower edge of filters at 
| least 3’-6" above fire 


BARBEQUE PITS Notes: 1. If hood is more than 12 feet 

long use muliiple takeoffs 6 

feet on center. 

. See VS—-S0-11 for information 
about filters and fans for 
range hoods. 


Q = 100 WH (maximum open door area, ft’) 
Minimum duct velocity = 1000 ~ 3000 fpm 

he = (filter resistance + 0.1") + 0.50 VP (straight take off) 
he = (filter resistance + 0.1°) + 0.25 VP (tapered take off) 


CHARCOAL BROILER AND 
BARBEQUE PIT VENTILATION : 


AMERICAN CONFERENCE 
f  =6OF GOVERNMENTAL 
J INDUSTRIAL HYGIENISTS 


FIGURE 


10.35 LABORATORY VENTILATION 


The primary method of contaminant control within the 
laboratory is exhaust ventilation and, in particular, laboratory 
hoods. This section presents information on laboratory hoods 
but expands to other types of ventilation control such as 
biological safety cabinets, clean benches, and other local 
exhaust systems found in the laboratory. 


10.35.1 Laboratory Hoods: In most cases, laboratory 
hoods will be purchased from manufacturers specializing in 
the design and construction of laboratory hoods. VS-35-01 
shows a typical laboratory hood design. VS-35-02 describes 
general-use laboratory hoods and VS-35-03 describes per- 
chloric acid hoods. VS-35-04 describes work practices for 
laboratory hoods. 


Several features are essential to the proper performance of 
the hood. The most important aspect of the hood is the 
aerodynamic entry characteristics. For the hood to adequately 
control contaminants, the entry must be smooth. This usually 
is achieved with an airfoil sill at the leading edge of the 
workbench. Often, beveled jambs at the side wall entry will 
improve the air flow. 


In many cases, good performance correlates with uniform 
face velocity. To achieve a uniform face velocity, many hood 
manufacturers provide adjustable slots in the plenum at the 
back of the hood. Although the adjustment will allow for 
unusual conditions such as large hot plates for sample diges- 
tions, inappropriate adjustment of the slots can have a detri- 
mental effect on hood performance.°35.) 


Supply Air Distribution: For typical operation of a labora- 
tory hood, the worker stands at the face of the hood and 
manipulates the apparatus in the hood. The indraft at the hood 
face creates eddy currents around the worker's body which 
can drag contaminants in the hood along the worker's body 
and up to the breathing zone. The higher the face velocity, the 
greater the eddy currents. For this reason, higher face veloci- 
ties do not result in greater protection as might be supposed. 


Room air currents have a large effect on the performance 
of the hood. Thus, the design of the room air supply distribu- 
tion system is as important in securing good hood perform- 
ance as the face velocity of the hood. American Society of 
Heating, Refrigeration and Air Conditioning (ASHRAE) re- 
search project RP-70 results, reported by Caplan and Knut- 
son,“°>2) conclude in part: 


1. Lower breathing zone concentrations can be attained 
at 50 cfm/ft? face velocities with good air supply 
distribution than at 150 cfm/ft? with poor air distribu- 
tion. With a good air supply system and tracer gas 
released at 8 liters per minute inside the hood, breath- 
ing zone concentrations can be kept below 0.1 ppm and 
usually below 0.05 ppm. 


2. The terminal throw velocity of supply air jets should 


Specific Operations 10-37 


be no more than one-half the hood face velocity; such 
terminal throw velocities are far less than conventional 
practice. 


3. Perforated ceiling panels provide a better supply sys- 
tem than grilles or ceiling diffusers in that the system 
design criteria are simpler and easier to apply, and 
precise adjustment of the fixtures is not required. 


For the reasons described, an increased hood face velocity 
may be self-defeating because the increased air volume han- 
dled through the room makes the low-velocity distribution of 
supply air more difficult. 


Selection of Hood Face Velocity. The interaction of supply 
air distribution and hood face velocity makes any blanket 
specification of hood face velocity inappropriate. Higher 
hood face velocities will be wasteful of energy and may 
provide no better or even poorer worker protection. The 
ANSI/ASHRAE Hood Performance Test‘'?*5*) may be used 
as a specification. The specified performance should be re- 
quired of both the hood manufacturer and the designer of the 
room air supply system. 


The specification takes the form: 


AUyyy, Alyyy, or AMyyy 


where: 
AU identifies an "as used" test 
Al identifies an "as installed" test 
AM identifies an "as manufactured" test 


yyy = control level, ppm, at the breathing zone of 
the worker. 


Any well-designed airfoil hood, properly balanced, can 
achieve < 0.10 ppm control level when the supply air distri- 
bution is good. Therefore, it would seem appropriate that the 
"AM" requirements would be < 0.10 ppm. The "AU" require- 
ment involves the design of the room supply system and the 
toxicity of the materials handled in the hood. The "AU" 
specification would be tailored to suit the needs of the labo- 
ratory room location. 


For projected new buildings, it is frequently necessary to 
estimate the cost of air conditioning early — before the 
detailed design and equipment specifications are available. 
For that early estimating, the guidelines listed in Table 10.35.1 
can be used. 


10.35.2 Biological Safety Cabinets: Biological safety 
cabinets (BSCs) are classified as Class I; Class II, Types A, 
BI, B2 and B3; and Class III. 


Class I BSC provides personnel and environmental protec- 
tion but does not protect the product. The front panel can be 
open, allowing room air to enter the cabinet, sweep the inner 
surfaces and exhaust out the duct. A front closure panel with 
glove ports may be installed. If gloves are installed, air is 


10-38 Industrial Ventilation 


TABLE 10.35.1. Laboratory Hood Ventilation Rates 


cfmift? 
Condition Open Hood Face 
1. Ceiling panels properly located with average panel face velocity < 40 fpm,(1035.2) Horizontal sliding sash hoods. No 60 
equipment in hood closer than 12 inches to face of hood. Hoods located away from doors and traffic ways.” 
2. Same as 1 above; some traffic past hoods. No equipment in hoods closer than 6 inches to face of hood. Hoods located 80 
away from doors and traffic ways.” 
3. Ceiling panels properly located with average panel face velocity < 60 fom MD Sth Ge ceiling diffusers properly located; 80 
no diffuser immediately in front of hoods; quadrant facing hood blocked; terminal throw velocity < 60 fpm. No 
equipment inhood closer than 6 inches to face of hood. Hoods located away from doors or traffic ways.” 
4. Same as 3 above; some traffic past hood. No equipment in hood closer than 6 inches to face of hood. 100 


5, Wall grilles are possible but not recommended for advance planning of new facilities 


*Hoods near doors are acceptable if 1) there is a second safe egress from the room; 2) traffic past hood is low; and 3) door is normally closed. 


drawn through a secondary opening equipped with a roughing 
filter. A laboratory hood, as shown in VS-35-20, could be 
considered a Class | BSC ifthe exhausted air is passed through 
HEPA filters prior to release to the atmosphere. 


Class If BSCs provide personnel, product, and environ- 
mental protection. Class II cabinets differ in the proportion of 
air recirculated within the cabinet; velocity of air flow to the 
work surface; where the exhausted air is discharged; and 
whether the contaminated air plenum is under positive pres- 
sure. A Type A cabinet (VS-35-10) may discharge the ex- 
hausted air, after HEPA filtration, directly into the room. Type 
A cabinets which discharge into the work area are not recom- 
mended for use with gases or vapors. A primary application 
is for sterile packaging. Care is required while decontaminat- 
ing the cabinet. 


Type B hoods (VS-35-11) discharge the exhaust but may 
recirculate within the cabinet. Type B1 cabinets recirculate 
about 30% of the air within the BSC and typically exhaust the 
remainder outside the laboratory (i.e., exhaust air is not dis- 
charged back into the room). The contaminated plenum is 
under negative pressure. Type B2 cabinets are referred to as 
"total exhaust" cabinets as the contaminated air is exhausted 
to the atmosphere after HEPA filtration without recirculation 
in the cabinet or return to the laboratory room air. Type B3 
BSCs have HEPA filtered downflow air that is a portion of 
the mixed downflow and inflow air from a common exhaust 
plenum. 


Class [Il BSCs (VS-35-20) provide the highest level of pro- 
tection to personnel and the environment. The cabinet is totally 
enclosed with operations conducted through attached gloves. See 
"National Sanitation Foundation Standard No. 49"°95) for 
descriptions and requirements of the various classes of BSCs. 


10.35.3 Clean Benches: Clean benches can be divided 
into laminar flow and exhausted clean benches. 


Laminar flow clean benches provide product protection 
only. In a laminar flow clean bench, room air is HEPA-fil- 
tered, directed across the work area and discharged back to 


the room. Air may be directed horizontally as depicted in 
VS-35-30 or vertically as in VS-35-3 1. Neither of these hoods 
provide worker protection. Workers using the Horizontal 
Laminar Flow Clean Bench are exposed to the product as the 
air sweeps across the product into the worker's face. Workers's 
arms or other objects protruding into the Vertical Laminar 
Flow Clean Bench opening may cause contaminated air to 
spill into the room. Personal protective equipment or general 
ventilation should be provided as needed. 


Other types of clean benches incorporate the same general 
principles of biological safety cabinets and utilize HEPA 
filtered laminar flow within the hood to provide product 
protection and exhaust sufficient air to ensure flow into the 
hood at the face to provide operator protection. 


10.35.4 Laboratory Equipment: Some laboratory equip- 
ment such as evaporation hoods (VS-35-40), discharge from 
instruments such as ICP or AA and some ovens (VS-35-41) 
require local exhaust ventilation to adequately contro! con- 
taminant releases. Often, specially designed ventilation spe- 
cific to the operation provides better control than using a 
laboratory hood to contro] these releases. 


REFERENCES 


10.35.1 Knutson, G.W.: Effect of Slot Position on Labora- 
tory Fume Hood Performance. Heating, Piping and 
Air Conditioning (February, 1984). 


10.35.2 Caplan, K.J.; Knutson, G.W.: Influence of Room Air 
Supply on Laboratory Hoods. American Industrial 
Hygiene Association Journal 43(10):738-746 
(1982). 


10.35.3 American Society of Heating, Refrigerating and Air 
Conditioning Engineers: ANSI/ASHRAE Standard 
110-1995, Method of Testing the Performance of 
Laboratory Fume Hoods. ASHRAE, Atlanta, GA 
(1995). 


10.35.4 National Sanitation Foundation: Standard 49, Class 
I] (Laminar Flow) Biohazard Cabinetry. NSF, Ann 


Specific Operations 10-39 


Arbor, MI (1987). Washington, DC (July 1963). 

10.35.5 U.S. Air Force: Technical Order 00-25-203: Stand- 10.35.6 Harris, W.P.; Christofano, E.E.; Lippman, M.: Com- 
ards and Guidelines for Design and Operation of bination Hot Plate and Hood for Multiple Beaker 
Clean Rooms and Clean Work Stations. Office of Evaporation. American Industrial Hygiene Associa- 


Technical Services, Department of Commerce, tion Journal 22(4) (August 1961). 


10-40 Industrial Ventilation 


~ Exhaust duct 


Optional room air 
by-pass does not — 
open until sash is 
closed 25-30 % 


/- Adjustable top slot 


— Sash closes by-pass 
when raised 


Air foil jamb — 


| — Fixed center slot 


Moveable sash ++ 
can have horizontal 
sliding panels 


“| | --- Rear baffle 


Recessed bottom — 
Airfoil sill —~ 


ae ee 


VERTICAL SASH AIRFOIL HOOD 


Q = 80-100 cfm/ff full open arec 
- For safety shield, at least depending on quality of supply 
one sash 16° max. width air distribution and uniformity 
of face velocity 
Ne = 0.5 VPy 
Duct velocity = 1000-2000 fpm to 
suit conditions 


Design specifications: 
General use laboratory hoods—See VS~45-02 : 


Perchloric acid —See VS~35-03 


"Auxiliary Air’ or “Compensating hoods 
furnish some replacement air at hood face, 
design varies with vendor. 


Work practices — See VS-—35-04 


HORIZONTAL SASH 
AIRFOIL HOOD 


AMERICAN CONFERENCE 
' OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


TYPICAL LABORATORY HOOD 


ia O22 91 | 


5-35 


FIGURE 


Specific Operations 10-41 


GENERAL USE LABORATORY HOODS 
A. Provide uniform exhaust air distribution in hood. Adjust baffles and air flow for < 10 percent 
variation in point-to-point face velocity with sash in maximum open position. 


B. Locate hood away from heavy traffic aisles and doorways. Hoods near doors are acceptable 
if: 1) there is a second safe means of egress from room, 2) traffic past hood is low, and 3) door 
is normally closed. 


C. Use corrosion-resistant materials suitable for expected use. 


D. Provide air cleaning on exhaust air if necessary and adequate stack height to minimize re-entry 
of contaminants or to comply with air pollution regulations. 


E. Avoid sharp corners at jambs and sill. Tapered or round hood inlets are desirable. An airfoil 
shroud at sill is important. 


F. Provide filters for radioactive materials in greater than “exempt” quantities. 


By-pass opening in hood is desirable to avoid excessive indraft under partially closed sash 
condition. Opening to be baffled to prevent splash from eruption in hood as shown in VS-35-01. 


H. Provide tempered or conditioned replacement air to laboratory. Replacement air volume to be 
selected for desired air balance with adjoining spaces. 


|. In order to reduce air flow volumes, local exhaust hood should be considered instead of 
laboratory bench hoods for fixed setups. 


J. For air conservation, use horizontal sliding sash with airfoil sill. 
All bench hoods should have a recessed work surface and airfoil sill. 


GENERAL USE LABORATORY 
HOODS 


as oT = = 


AMERICAN CONFERENCE 
/ OF GOVERNMENTAL 
: INDUSTRIAL HYGIENISTS 


10-42 Industrial Ventilation 


PERCHLORIC ACID HOODS 
Perchioric acid is extremely dangerous because it is a very strong oxidizer. When the acid reacts 
with organic material, an explosive reaction product may be formed. 


1. Donotuse perchloric acid in a hood designed for other purposes. Identify perchloric acid hoods 
with large warning signs. 


Provide exhaust ventilation and room supply air with minimal challenge to the hood. 


3. Utilize local exhaust ventilation within the hood to minimize condensation of vapors inside the 
hood. 


Locate all utility controls outside the hood. 


Materials of construction for this type of hood and duct must be nonreactive, acid resistant, and 
relatively impervious. AVOID ORGANIC MATERIALS unless known to be safe. Stainless steel 
type 316 with welded joints is preferred. Unplasticized polyvinyl chloride or an inorganic ceramic 
coating, such as porcelain, is acceptable. 


6. Ease of cleanliness is paramount. Use stainless steel with accessible rounded corners and 
all-welded construction. 


7. The work surface should be water tight with a minimum of 0.5-inch dished front and sides and 
an integral trough at the rear to collect the washdown water. 


8. Design washdown facilities into the hood and duct. Use daily or more often to thoroughly clean 
perchloric acid from the exhaust system surfaces. 


9. Each perchloric acid hood should have an individual exhaust system. Slope horizontal runs to 
drain. Avoid sharp turns. 


10. Construct the hood and duct to allow easy visual inspection. 


11. Where required, use a high-efficiency (greater than 80%) wet collector constructed for per- 
chloric acid service. Locate as close to the hood as possible to minimize the accumulation of 
perchloric acid in the exhaust duct. 


12. Use only an acid-resistant metallic fan protected by an inorganic coating or an air injector. 
13. Lubricate the fan with a fluorocarbon-type grease. 
14. Locate the fan outside the building. 


15. The exhaust discharge must terminate out-of-doors, preferably using a vertical discharge cap 
that extends well above the roof eddy zone. See Figure 5.30. 


PERCHLORIC ACID 
| HOOD NOTES 
parE O2— 9] rieure Vs— 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
DUSTRIAL HYGIENISTS 


a) E 


Specific Operations 


WORK PRACTICES FOR LABORATORY HOODS 


No large, open-face hood with a low face velocity can provide complete safety for a worker standing 
at the face against ail events that may occur in the hood. The hood may not adequately protect the 
worker from volatile or otherwise airborne contaminants with a TLV in the low part-per-billion range. 


For 


more ordinary exposures, a properly designed hood in a properly ventilated room can provide 


adequate protection. However, certain work practices are necessary for the hood to perform 
capably. The following work practices are generally required; more stringent practices may be 
necessary in some circumstances. 


1. 


Conduct all operations that may generate air contaminants at or above the appropriate TLV 
inside a hood. 


Keep all apparatus at least 6 inches back from the face of the hood. A stripe on the bench 
surface is a good reminder. 


Do not put your head in the hood when contaminants are being generated. 


Do not use the hood as a waste disposal mechanism except for very small quantities of volatile 
materials. 


Do not store chemicals or apparatus in the hood. Store hazardous chemicals in an approved 
safety cabinet. 


Keep the hood sash closed as much as possible. 
Keep the slots in the hood baffle free of obstruction by apparatus or containers. 
Minimize foot traffic past the face of the hood. 


Keep laboratory doors closed (exception: some laboratory designs require lab doors to be 
open). 


Do not remove hood sash or panels except when necessary for apparatus set-up; replace sash 
or panels before operating. 


Do not place electrical receptacles or other spark sources inside the hood when flammable 
liquids or gases are present. No permanent electrical receptacles are permitted in the hood. 


Use an appropriate barricade if there is a chance of explosion or eruption. 


Provide adequate maintenance for the hood exhaust system and the building supply system. 
Use static pressure gauges on the hood throat, across any filters in the exhaust system, or 
other appropriate indicators to ensure that exhaust flow is appropriate. 


If hood sash is supposed to be partially closed for the operation, the hood should be so labeled 
and the appropriate closure point clearly indicated. 


10-43 


AMERICAN CONFERENCE J WORK PRACTICES FOR 


Of 
INDU 


' GOVERNMENTAL ! LABORATORY HOODS 


STRIAL HYGIENISTS | 


10-44 Industrial Ventilation 


rr tt 
tv ft 
ff eee 


HEPA exhaust 


filter in Le in 

palceeine eT 

nee fs Ae Perea Supply filter 
access 

HEPA supply i O 

filter dl Jt HE ul | 

o View screen 

Vertical laminar 

Air flow 70-100 fpm fl Jl di \ Work zone 
with solid 
work surface 

= CL 

ositive 

pressure a -—— work opening 


plenum Air flow 80-100 fpm 


Exhaust filter 
cess 


NN 


Blewer Negative pressure plenum 


CLASS Il TYPE A 


| AMERICAN CONFERENCE | BIOLOGICAL SAFETY CABINET | 
: OF GOVERNMENTAL Chess i Tyee A : 


Specific Operations 10-45 


filter 


—HEPA = HEPA 
supply filter supply 


filter 


Vertical 
laminar 
Bair flow pop xe air flow “7p 
| 70-100 70-100 
A fpm 


| Vertical 
aaron laminar 


fpm . 
Air flow 
sfork\ lie J frork\ free 


80-100 
ork ia ZONNG a fon 


RECIRCULATING AIR HOOD * 100% EXHAUST AIR HOOD 


* Recirculating Air Hoods are not recommended for use with gases or vapors. 


Note: See “National Sanitation Foundation Standard 49” (10.35.4) for 
requirements and definitions of classes. 


For product protection only, see VS—35-30 and VS-35-31. 


AME RIC AN CON FE ERE NCE | BIOLOGICAL SAFETY CABINET 
OF GOVERNMENTAL — | CLASS. dh. TYPE. # 


10-46 Industrial Ventilation 


To HEPA filter =~ 
\ 


\ 
Roughing )} 
Glass | filter 
| window 


Hho closing] | 
1 door F 


- Glove 


Q = 50 cfm/ft? of open door area and 0.25” wg SP 
on a closed system. 
he = 0.50 VP, 
Duct velocity = 2000-4000 fpm 
Filters : 1. Inlet air filters in doors. 
2. Roughing filter at exhaust connection fo hood. 
3. HEPA filter 
All facilities totally enclosed in hood. Exterior controls may be advisable. 
Arm length rubber gloves are sealed to glove port rings. 
Strippable plastic on interior and air cleaner on exhaust outlef may be 
used fo facilitate decontamination of the system. 
Filter units may be installed in the doors to allow the air flow necessary 
for burners etc. 
For filters, see Chapter 4. 


P AMERICAN CONFERENCE |. DRY BOX OR GLOVE HOOD 


| FOR HIGH TOXICITY & RADIOACTIVE] 
MATERIALS 
bate 29] _~*pOuRE VS 3520) | 


OF GOVERNMENTAL 
PDOs HYG! Nis iis 


Specific Operations 10-47 


—_ HEPA filters 


\ _ Sides, top and rear 
| / __/ flush with filter outer 


foxered ood JA edges 


Nt ce 


eae 
ae 


COPS 
3 
oe 


PRK Ke 
so ecareuas 
ieee 

eaten 


intake 


BI ree Go 
ower ow i E: om 
Ceres. ve 


A cheat 


Face velocity = 90 fpm + 20 fpm 


Note : Total power input must be considered 
as part of air conditioning load. 


This hood does not provide protection 
for the operator. 


Reference 10.55.5 


HORIZONTAL LAMINAR FLOW 
CLEAN BENCH 
( TY 


ICAN CONFERENCE 
GOVERNMENTAL 


AMER 
OF 


Ba ee A APRODUCT PROTECTION “ONLY! 1 
} INDUSTRIAL HYGIENISTS EE ——9S—o7—— rom Teas OT 


10-48 Industrial Ventilation 


— Blower 


filter 


Roughing filter —~_ 
if needed - 


Sliding glass door 


Grating work area 


—~ Exhaust 


Blower located outside ce = : ’ 
Treat to suit contaminant 


cabinet 


Vertical velocity = 90 fpm with average minimum 
uniformity + 20 fpm 
Duct velocity = 2000 — 4000 fpm to suit conditions 


Clean station for control of air particles 


Supply and exhaust should be maintained equal by 
flow meter control techniques. 


This hood does not provide protection for the operator. 


Do not use with toxic material. 


VERTICAL LAMINAR FLOW 
CLEAN BENCH 


AMERICAN CONFERENCE 
_ OF GOVERNMENTAL = | (propucT PROTECTION ONLY) 
INDUSTRIAL HYGIENISTS fRE—go—g——pReme os 


Specific Operations 


Shel - Size plenum for —~- Size holes or slots 
1000 fpm down flow for 2000 fpm 


Strip heaters built — : GAS IER PEREE 


into bench 2 \ isl a a _ Height to suit glassware 


Z —|— ue 


Z| 


EVAPORATION BENCH 


Q= 20 cfm/foot of hood or 50 HL 
Minimum duct velocity = 2000 fpm 


he = 1.78VP, + 0.25VPy 


Strip heaters built — J 
into shelves 


EVAPORATION HOOD 


Q = 20 cfm/foot of shelf or 50 HL for each shelf 
Minimum duct velocity = 2000 fpm 
he = 1.78VP, + O.25VP, 


erence 10.35.6 


OF GOVERNMENTAL HOOD DESIGNS 
NDUSTRIAL HYGIENISTS are el ORE Oe OS 


10-49 


MERICAN CONFERENCE SPECIALIZED LABORATOF 


ae 


10-50 Industrial Ventilation 


Oven (dashed line) 


Skirts on sides 
and back desirable 


Maximum 2” gap I 


os ——~ Oyen Door 


Q = 200 — 400 cfm 
Minimum duct velocity 1000 -— 3000 fpm 


es 0.25 VEY 


p AMERICAN CONFERENCE 
/ OF GOVERNMENTAL 
: INDUSTRIAL HYGIENISTS [ar 


OVEN BXHAUST 


10.40 LOW VOLUME-HIGH VELOCITY EXHAUST 
SYSTEMS 


The low volume-high velocity (LVHV) exhaust system is 
a unique application of exhaust which uses small volumes of 
air at high velocities to control dust from portable hand tools 
and machining operations. Control is achieved by exhausting 
the air directly at the point of dust generation using close-fit- 
ting, custom-made hoods. Capture velocities are relatively 
high but the exhaust volume is low due to the small distance 
required. For flexibility, small diameter, light-weight plastic 
hoses are used with portable tools. This results in very high 
duct velocities but allows the application of local exhaust 
ventilation to portable tools which otherwise would require 
larger flow rates and large duct sizes when controlled by 
conventional exhaust methods. The resulting additional bene- 
fit is the reduction of replacement air requirements. 


This technique has found a variety of applications although 
its use is not common. Rock drilling dust has been controlled 
by using hollow core drill steel with suitable exhaust holes in 
the drill bits. Air is exhausted either by a multi-stage turbine 
of the size generally used in industrial vacuum cleaners or, in 
the case of one manufacturer,“°*°) by the exhaust air from 
the pneumatic tool which operates a Venturi to withdraw air 
from the drill. Some applications use flexible connections to 
a central vacuum system to aid in the control of graphite dust 
at conventional machining operations. One- to two-inch di- 
ameter flexible hose was used with simple exhaust hoods 
mounted directly at the cutting tool. In a similar application 
for the machining of beryIltum,(°*°” a central vacuum system 
utilizing 1.5-inch I.D. flexible hoses was employed. The 
exhaust hoods were made of lucite or transparent material and 
were tailor-made to surround the cutting tools and much of 
the work. Exhaust flow rates vary from 120-150 cfm with 
inlet velocities of 11,000-14,000 fpm. In another applica- 
tion,“°4°) a portable orbital sanding machine has been fitted 
with a small exhaust duct surrounding the edge of the plate. 
A fitting has been provided to connect this to the flexible hose 
of a standard domestic vacuum cleaner. 


VS-40-01 to VS-40-06 illustrate a custom-made line of 
exhaust hoods available.“°*°% The required air flow rates 
range from 60 cfm for pneumatic chisels to 380 cfm for swing 
grinders. Due to the high entering velocities involved, static 
pressures are in the range of 7" to 14" of mercury (95 to 290 
"wg). This high pressure is necessary to create the high 
capture velocities at the source to control the dust. However, 
there are disadvantages associated with high velocities: 1) 
small metal parts can be sucked into the hood; 2) coolants may 


Specific Operations 10-51 


be disturbed; and 3) very high noise levels may be produced. 


10.40.1 Design—Calculations: With the exception of the 
proprietary system mentioned which can be purchased as a 
"package," the design calculations for these systems are 
largely empirical and little performance data are available for 
the user. In normal ventilation practice, air is considered to be 
incompressible since static pressures vary only slightly from 
atmospheric pressure. However, in LVHV systems the ex- 
treme pressures required introduce problems. of air density, 
compressibility, and viscosity which are not easily solved. 
Also, pressure drop data for small diameter pipe, especially 
flexible tubing, is not commonly available. For practical 
purposes, the turbine exhauster should be selected for the 
maximum simultaneous exhaust flow rate required. Resis- 
tance in the pipe should be kept as low as possible; flexible 
tubing of less than |- to 1.5-inch diameter should be limited 
to 10 feet or less. In most applications, this is not a severe 
problem. 


The main consideration in piping for such systems is to 
provide smooth internal configuration so as to reduce pressure 
loss at the high velocities involved and to minimize abrasion. 
Ordinary pipe with threaded fittings is to be avoided because 
the lip of the pipe or male fitting, being of smaller diameter 
than the female thread, presents a discontinuity which in- 
creases pressure loss and may be a point of rapid abrasion. 


For dust exhaust systems, a good dust collector and primary 
separator should be mounted ahead of the exhauster to mini- 
mize erosion of the precision blades and subsequent loss in 
performance. Final balance of the system can be achieved by 
varying the length and diameters of the small flexible hoses. 

It must be emphasized that although data are empirical, 
LVHV systems require the same careful design as the more 
conventional ones. Abrupt changes of direction, expansions, 
and contractions must be avoided, and care must always be 
taken to minimize pressure losses. 


REFERENCES 


10.40.1 Thor Power Tool Company, Aurora, IL. 


10.40.2 Chamberlin, Richard I.: The Control of Beryllium 
Machining Operations. American Medical Associa- 
tion. Archives of Industrial Health, Vol. 19, No. 2 
(February 1959). 


10.40.3 Master Power, Inc., Westminster, MD. 


10.40.4 Hoffman Air and Filtration Div., Clarkson Indus- 
tries, Inc., New York. 


10-52 Industrial Ventilation 


-Annular 
slot 


Cone wheel used for} 


i 6 7 - 14 "Hg internal grinding on Ff 
' (0) castings and dies H 


Qn 
Branch static pressure = 7 to 14°Hg 
Slot velocity = 24,000 to 39,000 fpm 
Flexible hose = 1° to 1 1/2 ID 
Extension hose = up to 6 ft long* 


Grinding wheel sizes = 1° to 3° diam. 
1° to 4° long 


Peripheral speed = 6,000 to 10,000 linear fpm 


*Hose lengths may be extended up to : \ 
a maximum of 50 ft using larger sizes \ 


between the tool hose and the tubing system. 


Reference 10.40.4 


| EXTRACTOR HEAD FOR CONE 
I\WHEELS AND MOUNTED POIN TS 
FIGURE 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


ar a CSTE 


Specific Operations 


7 Adapter plate 
to fit grinder 


Hood adjustable for wheel wear 


i Hood 
a AL LLEL AVL ET LALLA LAP TEA ELT : 
j 


~ Minimum clearance 


Q = 25-60 cfm/inch diam. or width 

Branch static pressure = 7 to 14 "Hg 

Slot velocity = 30,000 to 39,000 fpm 

Flexible hose = 1° to 2” ID 

Extension hose = Up to 8 ft long* 

Peripheral speed = 6,000 to 12,000 linear fpm 


*Hose lengths may be extended up to a 
maximum of SO ft by using larger sizes 
between the tool hose and the tubing 


system. Hood fitted to grinder 


Reference 10.40.4 


10-53 


f AMERICAN CONFERENCE | oe ] pe ae eer : 
OF GOVERNMENTAL — | ee 


| INDUSTRIAL HYGIENISTS | 


10-54 Industrial Ventilation 


¥ ~Q y : Dust is extracted through slots builf into 
KN NY the rubber sleeve; slots should be on both 
o sides of the cutting edge of the chisel 


Q = 25-60 cfm/inch diam. 


Branch static pressure = 7 to 14 “Hg a = an Slots 
Slot velocity = 24,000 to 39,000 fpm Jie CB pee 


Flexible hose = 1 to 1 1/2” ID 


Extension hose = Up to 8 ft long* y i | 
Chisel sizes = 13/16 octagonal iy la | | | 
7/8 octagonal if 

7/8 hexagonal | | 3 fe / /\ 
*Hose lengths may be extended up to ANG ” /, \ 
a maximum of 50 ft by using larger z | is 


sizes between the tool and the tubing i : \ 
system. ae 


| Reference 1040.4 
AMERICAN CONFERENCE 
| OF GOVERNMENTAL 
INDUSTRIAL HYGIENIS 


PNEUMATIC CHISLL SLEEVE 


my Soeewes 
r's FIGURE 


Specific Operations 10-55 


an 
N 
Adjustable 


= a 
c= 


“7-14 "Hg 


These extractor heads have been specifically designed 

for work done inside casting or in awkward places 

when radial wheels of small diameter are most suitable. 
The heads are narrower than the grinding wheels and can 
precede the wheel when a groove is being ground. 


s—~ Peripheral dust captured 
v~ Fine dust controlled 


yo Heavy particles 


Q = 70-150 cfm/inch diam. 

Branch static pressure = 7 to 14” Hg 
Slot velocity = 25,000 to 39,000 fpm 
Flexible hose = 1° to 1 1/2 JD 
Extension hose = up to 8 ft long* 


Grinding wheel sizes = 2° diam. x1/2” 
to 8° diam. x 2” wide = 


~ 


Peripheral speed = 6,000 to 15,000 linear fpm ™ 


*Hose lengths may be extended 
up to a maximum of 50 ft by 
using larger sizes between the 


tool h d the tubi term. 
EOP MIRON) oubeet Mme ctenneay eon Fe Reference 10.40.4 


AMERICAN CONFERENCE | EXTRACTOR HEAD FOR SMALL 
OF GOVERNMENTAL RADIAL GRINDERS 


| INDUSTRIAL HYGIENISTS Fx: op ee ' 


10-56 Industrial Ventilation 


Q= 
Branch static pressure = 7 to 14 "Hg 
Slot velocity = 24,000 to 39,000 fpm 
Flexible hose = 1° to 2° ID 

Extension hose = Up to 8 ft long* 


Sanding disc size = 2° to 9” diam. 
Peripheral speed = 4,500 — 14,000 linear fpm 


*Hose lengths may be extended up to a 
maximum of 50 ft by using 
larger sizes between the tool 
hose and the tubing system. 


Reference 10.40.4 


AMERICAN CONFERENCE J 
OF GOVERNMENTAL | 


: INDUSTRIAL HYGIENISTS 


EXTRACTOR HOOD FOR 
DISC SANDER 


SED Teen ESTEE : 


Specific Operations 10-57 


This design is suitable for sanders 
running up to 20,000 cycles per minute 


Q = 5 to 15 cfm/inch of perimeter 
Branch static pressure = 7 to 14° Hg 
Slot velocity = 15,000 to 39,000 fpm 
Flexible hose = 1 1/4" to 2” ID 
Extension hose = Up to 8 ft long* 


*Hose lengths may be extended up to a 
maximum of 50 ft by using larger sizes 
between the tool hose and the tubing 

system. 


| AMERICAN CONFERENCE | 
f OF GOVERNMENTAL 


EXTRACTOR TOOL FOR | 
VIBRATORY SANDER : 


| INDUSTRIAL HYGIENISTS mE TOL OO RO STAG 


10-58 Industrial Ventilation 


System Notes 
Bell and socket, smooth—fiow type 
tubing and fittings should be used 
throughout the system. 
When system is used for vacuum 
cleaning of abrasive materials, Schedule To Atmosphere 
No. 40 pipe and cast iron drainage fittings, 


or heavier, should be used in place of 4 


I.D. Plastic 
Hose Size 
(inches) | 


Disk sanders 3-9 inch diam. |60-175; 1-1.5 
paren pad sander — 4°x9" | 1.25 
Router, 1/8 -1 

| Belt sander 3°-4000 fpm 
i Pneumatic chisel 

Radial wheel grinder ~ Automatic Bag Cleaner 
Surface die grinder, 1/4” (Recommended) 

Cone wheel grinder oe 


Cup stone grinder, 4 ; \ 
Cup type brush, 6” : Th «—- Dust collector 


Radial wire brush, 6” 
Hand wire brush 3° x 
Rip out knife 

Rip out cast cutter 
Saber saw ; 
Swing frame grinder 2° x 18° | . v~ Primary 
oe abrasive: Ea : ; separator 


=e Pa (im 


1/8” I} 2 1/3” 


#7” -72,000 rpm Chipping 6’ —-10,000 rom 6 x1", 10,000 rpm 
g Disc sander hammer Cup stone grinder Wheel grinder 


Swing frame 
grinder , 
Reference 10.40.4 


Saal CONFERENCE IVPICAL oYslEM. 
GOVERNMENTAL LOW VOLUME HIGH VELOCITY J 


INDUSTRIAL HYGIENISTS mE ODO EUR VSD 


10.45 MACHINING 


The primary purpose of metal cutting machines is to finish rough 
parts formed by other processes to specific dimensions. Finishing 
and shaping may be accomplished by a variety of cutting tools such 
as saws, broaches, and chisel-shaped tool bits held in fixtures with 
fixed or movable drives. Cutting is accomplished by creating 
relative motion between the tool edge and the material blank. Chips 
of varying sizes are produced; chip size depends on the material 
being cut, feed rate of the tool, and relative speed or feed between 
the tool and the metal being shaped. 


Non-traditional methods of metal cutting and shaping in- 
clude electrochemical, electrodischarge, wire electrodis- 
charge, and laser beam machining. With the exception of the 
laser beam, each of the processes utilizes a circulating oil or 
water-base dielectric to facilitate molecular erosion as well as 
to remove process heat and particulate debris. The rate of 
metal removal is controlled carefully by regulating the flow 
of electrical current between the shaped anode or wire and the 
workpiece. The laser beam is used in a dry environment and 
metal cutting is accomplished by vaporizing the workpiece 
along the cutting edge with a focused beam of high energy 
light. The process is flexible and a variety of metallic and 
non-metallic materials can be shaped by this means.(04°.) 


It is estimated that up to 97% of the work involved in 
conventional metal cutting results in heat. The rate of heat 
removal must be controlled carefully in order to protect both 
the cutting tool and the metallurgy of the work being cut. 
Where convection or radiant cooling is insufficient, a cutting 
fluid can be used to reduce friction, carry away generated heat, 
and, more commonly, flush away metal chips produced by the 
cutting process. Cutting fluids include straight-chained and 
synthetic mineral oils as well as soluble oil emulsions in 
water. A variety of water soluble lubri-coolants (1~5% mix- 
ture of lubricants, emulsifiers, rust inhibitors, and other 
chemicals in water) are used commonly, particularly for high 
speed metal working machines. In some applications the 
lubri-coolant mixture is applied as a mist by using a small 
volume of liquid in a high velocity air stream. In the more 
usual situation, liquid is applied by flooding the tool in the 
cutting zone to flush away cutting debris. The latter type 
system requires a low pressure pump with valves; filters; 
settling chamber to separate the fluid from the chips; and a 
reservoir which permits recirculation. Where liquids cannot 
be used, low temperature nitrogen or carbon dioxide gas can 
be used as a cooling media for both the tool and the cutting 
surface as well as a means of dispersing particulate debris. 


The hazards created by skin exposure to the water lubri-cool- 
ant mixtures, particulates and oil mist/vapor produced in the 
transfer of heat is best handled with engineering controls—pri- 
marily ventilation. An additional health concern is the fact that 
soluble oil emulsions provide a breeding ground for bacteria and, 
therefore, it is common practice to add biocides to prevent odor 
generation and decomposition of the oil mixture. Biocides and 


Specific Operations 10-59 


other additives may be primary skin irritants or cause hyper- 
sensitive dermatitis. It is for these reasons that mist, vapors, 
and particulates must be controlled adequately.“'°*”) 


Mist and vapors from machining operations can be control- 
led by a combination of machine enclosure and local exhaust 
ventilation. Exhaust hoods and enclosures should be designed 
so the machine can be serviced easily and the operation 
observed when required. Hood sides should act as splash 
guards since an indraft of air will not stop liquid directly 
thrown from rotating parts. All components should be robust 
and rigidly supported. To facilitate maintenance, service and 
tool adjustment, portions of the hood enclosure which are not 
permanently fixed should be designed for easy removal. 
Thought should be given also to the use of sliding, hinged, or 
bellows-connected panels in locations where frequent access 
is required. All windowed openings must be shatter-proof 
with appropriate internal lighting. Allnon-fixed panels should 
be designed with overlapping, drip-proof edges. The use of 
gaskets or seals on abutting panels is not recommended. 
Ventilation rates vary; however, a minimum of 100 fpm 
indraft usually is required to prevent vapor and mist from 
exiting the enclosure. A typical machine enclosure will re- 
quire a volumetric flow rate of from 400-500 cfm minute. 
Additional air may be required to control heat generated 
within the enclosure as well as to maintain adequate vision. 
Where coolant flumes are used for chip transport, additional 
exhaust ventilation is required to control air entrainment. 
Baffles above the liquid level are beneficial and flumes should 
be enclosed to the extent possible. 


Local ventilation control is the preferred method—particu- 
larly in machine environments which are temperature control- 
led with refrigerated air conditioning systems. In more open 
workrooms, the use of dilution ventilation may be adequate 
to control air contaminants. For further information on dilu- 
tion ventilation, see Chapter 2 of this manual. 


REFERENCES 


10.45.1 Rain, Carl: Non-traditional Methods Advance Ma- 
chining Industry. High Technology (November/De- 
cember 1957). 


10.45.2 O'Brien, Dennis; Frede, John C.: Guidelines for the 
Control of Exposure to Metal Working Fluids. Na- 
tional Institute for Occupational Safety and Health 
(February 1978). 


10.45.3 Schulte, H.F.; Hyatt, E.C.; Smith, Jr., F.S.: Exhaust 
Ventilation for Machine Tools Used on Materials of 
High Toxicity. American Medical Association Ar- 
chives of Industrial Hygiene and Occupational 
Medicine, Vol. 5, No. 2] (January 1952). 


10.45.4 Mitchell, R.N.; Hyatt, E.C.: Beryllium—Hazard 
Evaluation and Control Covering a Five-Year Study. 
American Industrial Hygiene Quarterly, Vol. 18, 
No. 3 (September 1957). 


10-60 Industrial Ventilation 


Booth sized—~ 
to suit 
work 


(= 


Q at booth = 225 cfm/ tt? open area 
Q at bottom = 350 cfm 


Minimum duct velocity = 4000 fpm 


he = 1.75 VPg, at point A 


ERIC 


| OF GOVERNMENTAL 


AN CONFERENCE | 
INDUSTRIAL HYGIENISTS 


METAL CUTTING BANDSAW 


TOE STIS OT TY 


Pare 10-90. 10290 > 3 


Specific Operations 10-61 


_i-- Transparent cover 
normally closed 


Q=300 cfm/ft*? of open area 
Minimum duct velocity = 3500 fpm 
he = 0.55 VPa 


Reference 10.45.4 


J AMERICAN CONFERENCE | HIGH TOXICITY MATERIALS | 
f «OF GOVERNMENTAL [> MILLING MACHINE HOOD 


10-62 industrial Ventilation 


slot in rear. 


haust 


Flange if possible. 


a~ 
EX 
if 4 
) i 
| 
| 
t 


DS 
Weer, 


—— Ex 


ey a (@ | 
! = ses say = aos 


slot, 
slot. 


x 
1 
2900 fom 


hrougn rear 


Through fron 


400 cfm/fi of shear length. 


40% 
60% 


i 
Minimum duct velocity 


lot velocity 


S 


AMERICAN CONFERENCE | METAL SHEARS 
OF GOVERNMENTAL | HIGH TOXICITY MATERIALS 


| INDUSTRIAL S BETO a | 


Revarence: tomes —— 


Specific Operations 10-63 


r Hinged die cover 


ee Parts container 


END VIEW SIDE VIEW 


header : 

O° = 750 ctm/ft of die opening 
Minimum duct velocity = 3500 fom 
hy = 1.0 VP, + 0.25 VP, 


Parts discharge and container : 


Q = 100 cfm/ft of hood length 
Minimum duct velocity = 3500 fom 
h, = 0.25 VP, 


J AMERICAN CONFERENCE COLD HEADING MACHINE 
f OF GOVERNMENTAL VENTILATION 
J INDUSTRIAL HYGIENISTS BETO To 


Industrial Ventilation 


10-64 


JuUD]OOD 4O} 
adid absnyosiq 


a 7 
ae Sale 


LS) 


LATHE HOOD 


Sjuod eZis winwixow 
4O}, |snsojoua aZIS 
4A OS'O = U4 

wd} QOSe = AjloojeA jonp WinuulUIW 
pain usdo jo OT ie oos =D 


st 
uw) 
t+ 
© 
o 
1S) 
Cc 
o 
~ | 
@ 
w 7G 
wo 
ow | 


pasojo Ajjpwuo0U 
4J8A09 jUsiDdsuDds] 


OF GOVERNMENTAL : 
F INDUSTRIAL HYGIENISTS [ing 760 TRO STAST OFT 


| AMERICAN CONFERENCE | 


10.50 MATERIAL TRANSPORT 


Ventilation of material transport systems generally requires 
the use of an exhausted enclosure because of the motion and 
quantity of material involved. If the enclosure were perfectly 
air tight there would be no need for exhaust. However, there 
usually are cracks and other leak points in addition to the 
openings necessary for personnel and material access. 


For enclosures where there is little motion and low material 
quantity, the exhaust rate is the product of the total openings 
in square feet and some velocity between 50 and 200 fpm. 
However, in some cases the inward flow of material and 
entrained air can overwhelm the exhaust flow rate calculated 
on the basis of enclosure openings. In such cases the material 
flow rate, the dustiness of the material, and the height of fall 
in transferring from one surface to another must be considered 
in the system design.(°*°!. !0502) Other design factors include: 


1. The rate of air induction into the space. 


2. The location of cracks or other openings in relation to 
the "splash" or agitation of material during transfer. 


3. The need to avoid excessive product withdrawal. 


4. Adequate air flow for dilution of interior concentra- 
tions for visibility or safety from explosions. 


10.50.1 Bucket Elevators: Air motion caused by the 
bucket moving within the elevator is not significant. The 
motion of buckets in one direction is offset by the opposite 
flow. Consequently, an exhaust rate of 100 cfm/ft? of elevator 
cross-section is adequate for most elevator applications (see 
VS-50-01 for details). Additional ventilation is required as 
materials enter and leave the elevator (see VS-50-10, VS-S0- 
20 and VS-50-21). Handling hot material often causes signifi- 
cant thermal buoyancy which requires increased exhaust 
ventilation to overcome this challenge. 


10.50.2 Conveyors: Dust from the operation of belt con- 
veyors originates mainly at the tail pulley where material is 
received and at the head pulley where material is discharged. 
The exhaust requirement at the head pulley is generally small 
because air is induced downward and away from this transfer 
point. An exhaust rate of 150-200 cfm/ft? of opening often is 
adequate. 


At the tail pulley, the exhaust requirements are determined 
by the amount of air induced by the delivery chute. An exhaust 
of 350 cfm/ft? of belt width often is adequate where the 


Specific Operations 10-65 


material does not fall more than 3 feet. The exhaust point 
should be located at least twice the belt width away from the 
point where the material hits the belt. Where the material falls 
more than 3 feet, additional exhaust is required (see VS-50-20 
for details). Note that very dry or dusty material may require 
flowrates 1.5 to 2.0 times these values. 


Belt conveyors should be covered and exhausted at 30 foot 
intervals at arate of 350 cfm/ft of belt width. Vibrating feeders 
should be exhausted at a rate of 500 cfm/ft of feeder width. 
Rubber or canvas flexible seals should be provided from the 
feeder sides and end to the hopper sides and end. 


The conveying of toxic material requires additional care in 
enclosure design to ensure that no air leaks out and that 
sufficient access is available for inspection and cleanout. The 
head pulley should be equipped with a scraper or brush (see 
VS-50-21). 


10.50.3 Bin and Hopper Ventilation: For the mechanical 
loading of bins and hoppers, the exhaust rates previously 
listed for belt conveyors are appropriate. An exhaust rate of 
150 cfm/ft? of hopper is adequate for manual loading opera- 
tions. The enclosure should cover as much of the hopper 
opening as possible. 


10.50.4 Loading and Unloading: For loading and un- 
loading operations, a ventilation rate of 150~200 cfm/ft’ of 
enclosure opening is adequate provided the enclosure is large 
enough to accommodate the "splash" effect. The entrance to 
enclosure for truck dumps should be covered with flaps to 
minimize ventilation requirements. Rotary or bottom car 
dumps generally are exhausted at the rate of 50-100 cfm/ft? 
of hopper area. 


REFERENCES 


10.50.1 DallaValle, J.M.: Exhaust Hoods. Industrial Press. 
New York (1946). 


10.50.2 Hemeon, W.C.L.: Plant and Process Ventilation. 
Industrial Press, New York (1963). 


10.50.3 Rajhans, G.S.; Bragg, G.M.: Engineering Aspects of 
Asbestos Dust Control. Ann Arbor Science Publica- 
tions, Inc. Ann Arbor, M1 (1978). 


10.50.4 National Grain and Feed Association: Dust Control 
for Grain Elevators. NGFA, Washington, DC 
(1981). 


10-66 Industrial Ventilation 


Alternate exhaust point 
oy 


Preferred exhaust 
/ point 


Additional ventilation for —7+ 
hopper, bin, or screen 

see Vs—50-10 & 

VS-99-01 


45°-60° 
eres ae 


Take-off detail 


& 


> 


y, Tight casing 


Additional ventilation 
for conveyor 
discharge. See below. 


\ 


For casing only 
Q =100 cfm/ft? casing cross section 
Minimum duct velocity = 3500 fpm 


he = 1.0 VPq or calculate from 
individual losses 


Take-off at top for hot materials, 
at top and bottom if elevator is over 
30 fi high, otherwise optional. 


CONVEYOR BELT DISCHARGE VENTILATION 


BELT SPEED 


FLOWRATE 


—t 


||Less than 200 fpm | 350 cfm/ft of belt width. Not less than 150 cfm/ft? of opening. 


: Over 200 fpm 500 cfm/ft of belt width. Not less than 200 cfm/tt of opening. 


VENTILATION 


AMERICAN CONFERENCE 
p OF GOVERNMENTAL 
f INDUSTRIAL HYGIENISTS 


Specific Operations 10-67 


Enclosed loading point 


i 
Locate remote from \ 
loading point — 


Minimum duct velocity = 3500 fpm 


Q = 200 ctm/ft of all open area 
he = 0.25 VFy 


MECHANICAL LOADING 


Belt speed 
Less than 200 fpm —-—--— 


Over 200 fpm 


Booth to accommodate 


Flowrate 


350 cfm/ft of belt width. 
Not less than 150 cfm/ft? of opening. 


500 cfm/ft of belt width. 
Not less than 200 cfm/ft* of opening. 


barrel, bag, “7 


Booth to cover as much | 
of hopper as possible § 


Minimum duct velocity 
150 cfm/ft? 


0.25 VR 


3500 fpm 


face 


MANUAL LOADING 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


BIN & HOPPER VENTILATION 


10-68 Industrial Ventilation 


| Elevator 
exhaust | 
e VS-50+01) 


Close face to eae ye 
bottom of belt ——. 


me NOTRE, | 
1. Conveyor transfer less As close as i” Tote box 
than 3’ fall. For greater practical -— 
fall, provide additional | 
exhaust at lower belt. ee 
See 3 below. 2. Conveyor fo elevator with 
hy, = 0,25. VPs magnetic separator. 
° Ng = O25 VE 4 


DESIGN DATA 


Transfer points: 


Enclose to provide 150 — 200 fpm indraft 
at all openings. (Underground mining 

ir a 1/3 tunnel ventilation will interfere with 
conveyor exhaust systems.) 


[ekeKedevorens) 


2’ clearance for load-—— 
on belt 


2 arene DETAIL OF BELT OPENING 


3. Chute to belt transfer and conveyor Q = 350 cfm/ft beit width for belt 
transfer, greater than 3° fall. speeds under 200 fpm. (minimum) 
Use additional exhaust at (A) = 500 cfm/fi belt width for bel 
for dusty material as follows: speeds over 200 fpm and for 
Belt width 12°-36', Q=700 cfm magnetic separators. (minimum) 
Belt width above 36°, Q=1000 cfm Minimum duct velocity = 3500 fpm 
Ne = 0.25 a h, = 0.25 VP, 


Conveyor belts: 


Note: Dry, very dusty materials may Cover belt between transfer points 
require exhaust flowrates 1.5 fo Exhaust at transfer points 
2.0 times stated values. Exhaust additional 350 cfm/ft. of belt 


width at 30’ intervals. Use 45 
tapered connections. 


AMERICAN CONFERENCE 
: OF GOVERNMENTAL 
INDUSTRIAL HYGIENIS 


CONVEYOR BELT VENTILATION 
JT TOT ROR VS 50-20 


TS 


Specific Operations 10-69 


ae Totally enclosed conveyor, 
| leakage factor depends on 


type of construction 
Settling can 


2 ae 
PL 


~~ Cleanoui and inspection 
doors 


Internal skirt board ~— 


Troughing belt - 


Scraper conveyor 


--- Return belt scraper 
or brush 


Q = 250 cfm/ft* of open area 
Minimum duct velocity = 3500 fpm 
he = 0.4 VP, 


: Reference: 10.50.3 


AMERICAN CONFERENCE | TOXIC MATERIAL BELT 


OF COVERNMENTAL CONVEYING HEAD PULLEY 
INDUSTRIAL HYGIENISTS FIGURE VS — 50 TT 


10-70 Industrial Ventilation 


— Square settling box 
(Designed for 150-200 fpm velocity) 


, OX 
NE 
eee Aa a 4 
| \ we | 


\-— Completely 
enclosed 
conveyor 


~ Clean-out and inspection 


Q = 250 ctm/ft” of open area 
Minimum duct velocity = 3500 fpm 
he = 0.4 VP, 


: Reference: 10.50.34 


CONVEYOR BELT LOADING 


AMERICAN CONFERENCE | TOXIC MATERIAL 


: OF GOVERNMENTAL 
NDUSTRIAL HYGIENISTS [Ra 


Specific Operations 10-7] 


Power hoist —-—~ 
| 


_7~ 8400 cfm 
we air suction 


Telescoping 
grain spout-—~ ~~ 


Telescoping : 


or flexible PSS 


4 ft. vertical 


a 


Sah 
Moveable hood 


with flexible _ 
skirt cna 


rea, ae - ’ 7 i i a 
a a H 5 ea 


a 


Reference 10.50.4 


| AMERICAN CONFERENCE 
f OF GOVERNMENTAL 
f INDUSTRIAL HYGIENISTS 


RAIL LOADING 


Date OO OT Se OL ~ PRIGURE Vs= 


50-50 | 


10-72 Industrial Ventilation 


LS 


A ae 
qua 
Air conduit 
= must be flex 
—— hose or swivel 
—— joint 


aN 
HVT 
} 


fev Telescoping 
grain spout 


~. 
SATE velocity 
1500 fpm 


Outer sleeve 
must be telescoping 
or collapsible 


ae 


Must maintain 
12 clearance 
to be most 

effective 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 3 
INDUSTRIAL HYGIENISTS | 


TRUCK LOADING 


10.55 METAL MELTING FURNACES 


This set of VS-prints describes hood designs for a variety 
of metal melting furnaces, including electric induction, carb- 
on arc, convention, and crucible, which use natural gas or 
electric resistance elements as the heat source. Exhaust ven- 
tilation usually is required to control specific oxides associ- 
ated with the metal being melted or contaminants carried in 
the scrap charge. In some cases, a single hood will suffice for 
charging, melting, and pouring. In other cases, a separate 
hood, remote from the primary melter, may be required for 
charging because of the nature of the charge or the large open 
area necessary at this phase of the operation. This is true 
particularly of electric arc furnaces which are completely 
open for charging. After charging, the port exhaust can be 
used to achieve control during the remainder of the melting 
and pouring cycles. 


All metal melting will produce a slag which must be 
removed prior to pouring. This activity may produce a signifi- 
cant release of oxides and may require a separate exhaust 
system for oxide and/or dross control. Where metal purifica- 
tion is performed directly within the furnace or melting vessel, 


Specific Operations 10-73 


such as the addition of oxygen or chlorine, additional exhaust 
may be required to contain the rapidly generated plume. 


All systems must be designed to include the increase in air 
temperature under operating conditions to insure an adequate 
air flow into the hood. The air temperature rise is usually 
relatively low except where metal innoculants or oxidizers are 
added to the molten charge. During this phase of metal 
melting, a significant temperature rise will occur and it is 
customary to provide a large hood in which gas expansion can 
take place. 


REFERENCES 


10.55.1 American Air Filter Co.: Rotoclone Dust Control 
(January 1946). 


10.55.2 Kane, J.M.: Foundry Ventilation. The Foundry (Feb- 
ruary and March 1946). 


10.55.3 Kane, J.M.: The Application of Local Exhaust Ven- 
tilation to Electric Melting Furnaces. In: Transac- 
tions of the American Foundrymen's Association, 
Vol. 52, p. 1351 (1945). 


10-74 Industrial Ventilation 


y - Crane track -- % 


\ 


yr Close end with panel ~ 


to ite 


7 Sliding panels 
ee on rollers ———~._ 


a or <a : f | “Track for panels- 
ease 1 : (ote ; 


Q = 200 inst of opening including doors, 
plus products of combustion corrected 
for temperature. Row of crucibles — 

Minimum duct velocity = 3500 fpm 


fo 


he = 0.5 VR Note: Same principle of sliding or swinging 
doors is applied to individual furnace 
enclosures. 


rm Canopy to clear crane; or provide slot for crane bridge; 
or separate cranes inside and outside; or use 
manual crucible removal. 


pocorn 


Zz 
200 cfm/ft of total opening, minimum, plus 
products of combustion corrected for temperature. 


MELTING FURNACE 
CRUCIBLE, NON-TILT 


pate 2-9] [ricurE VS—55—O01 , 


AMERICAN CONFERENCE 
p OOF GOVERNMENTAL 
f INDUSTRIAL HYGIENISTS 


Specific Operations 10-75 


coe —Door guides ————-----—---- 


. Counterweighted or ee ee Solid side 
spring~loaded sliding panels 
doors front and back if 
if necessary ——~__ 


fo A] ~ Door to extend below 
ri top of furnace if 
possible. 


Q = 200 LW; but not Jess than 
200 cfm/ft of all openings 
with doors open. Correct for products 
of combustion and temperature. 
Minimum duct velocity = 3500 fpm 
he = 0.25 VPy 


AMERICAN CONFERENCE MELTING FURNACE 
OF GOVERNMENTAL : TET ENG 
INDUSTRIAL HYGIENISTS 


10-76 Industrial Ventilation 


- Flanged by-pass connection; no exhaust 
during furnace tilting and pouring 


Hood is fastened 
to furnace roof 
and swings with 
roof —--—~ 


eal! 


7 ay 


Furnace body Al 


Close Capture: 


For Q, SP and operating temperature, consult manufacturer 
Approximate exhaust rate = 2500 scfm/ton of charge 


Alternate designs: 


1. Some exhaust designs utilize direct furnace roof tap. For details 
consult manufacturer. 


. Canopy hoods require large exhaustand are not recommended 


Canopy hoods can be used as secondary hoods to capture 
fugitive emissions. 


| References 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 


STRIAL HYGIENISTS 


10.55.2, 10.55.3 


10.55.1, 


MELTING FURNACE 


| ALECTRIC, TOP ELECTRODE | 
pee 02g er re 0s | 


LN 


Specific Operations 10-77 


Ly | | aot aes 
| \\ | oe sem tems oor <a = | 
| | ee oo eee 
. 2 -- Hinged an 
door | 
Rollers _s — Pe i | | 
Fe Steg lhe ge feck | | Abe 


> 
<a ae ee 
H ~~ XY 


| 
Sa 


Q = 400 eae of opening 
Minimum duct velocity = 3500 fpm 
he = 1.78 VR + 0.25 VP, 


e 


MELTING FURNACE’ 


| AMERICAN CONFERENCE | 
7 OF GOVERNMENTAL ELECTRIC ROCKING 


' INDUSTRIAL HYGIENISTS E—— eo oe ee + 


10-78 Industrial Ventilation 


100-200 cfm/ft of Co NOTE: Separate flue required if com— 
bustion gases are not vented 


opening plus volume of 
through the hoods. 


products of combustion.®* | 
Duct velocity = 3500 fpm | 
he = 0.50 VP, 


*Correct for temperature. 
See Chapter 3 

for additional 
information. 


~~ Work openings. 
Keep as small as practical. 
Doors advisable. 


Woe pot 


~Door for dross pan 
removal. 


Pes Dross chute, min angle = 60 


STATIONARY FURNACE OR MELTING POT 


AMERICAN CONFERENCE | MELTING POT & FURNACE 
f OF GOVERNMENTAL = [ 
| INDUSTRIAL HYGIENISTS REVS SST 


Sa a 


Specific Operations 10-79 


Flange is necessary — 


180° slot if possible — 
cy \ 


~ Flange 


Furnace \ A 
\. Size for 1/2 
slot velocity 


Furnace 
bottom 


Q= 175 cfim/ft” of furnace top with curved slot 
and flanges. 

Slot velocity= 2000 fpm 

Minimum duct velocity= 3500 fpm 

Entry losses= 1.78 VP, + 0.25 VPg 


AMERICAN CONFERENCE | CRUCIBLE MELTING PURNACE 
OF eee teens | HIGH eeae [T a MATERIAL 


es TT S506 | 


10-80 Industrial Ventilation 


Charging door 


Electrical cables 
S|) 


and cooling Pouring channel 

lines : 
| 
l 
l 
l 
l 


| 
=: | 
\ | T] 
| | 
_Furnace) Rotating joint I Jee | 


FRONT VIEW SIDE VIEW 


Rotating aa 


TOP VIEW 


ISOMETRIC VIEW 


Q = 350 cfm /tt? open area. Correct for 
temperature and combustion products. 


Entry loss = 1.78 VP5 + 0.5 VFy 
Slot velocity = 2000 fpm 
Minimum duct velocity = 3500 fpm 


AMERICAN CONFERENCE | INDUCTION MELTING FURNACE 
| OF GOVERNMENTAL TILTING 


| INDUSTRIAL HYGIENISTS fgg —75—97 REVS SSO 


Specific Operations 10-81 


Use top baffle when 
operations permit | > 


NSS T 
Vie Xs min 
[7 Wide flange 


paz ce re ‘aaa iid Xa | 


F ZLZ LEAL. oe 
pee eB W sai, 


Q = 200 (10 x? + A) where A equals hood area. 
Minimum duct velocity = 3500 fpm 

he= 1.78 VPs + 0.25 VPy 

Use slots for uniform distribution, size slots 

for 2000 fpm 


No 


K~ 


PARTIAL SIDE ENCLOSURE 
ae 


Pd 
p— SJ 


pe 


—-——— Use slots for distribution 


Slot velocity = 2000 fpm 


# 
r 


Note: 
For large molds and ladles 
provide large — draft hood 
similar to shakeout. 

Q = 400 cfm/ft* working area. 


cfm/ft of hood ijength. 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


POURING STATION 


| SEIT 


10-82 Industrial Ventilation 


Flange type fitting for 
easy removal of hood 
(if necessary) -——---» 


- Hinged baffle for preventing short 
circuiting of air. 


Place hood as close to machine as 
possible. If more than 4 inches 
from back of machine, hinged side 
baffles should be used. 


Q = 300WH 
he= 0.25 VPy 


Minimum duct velocity = 


AMERICAN CONFERENCE 
OF GOVERNMENTAL : 
INDUSTRIAL HYGIENISTS Ry 


Products of combustion 
require separate flue or 
may be vented into hood. 


2000 fpm. 


FIXED POSITION 
DIE CASTING HOOD 


pricuRE YS AS 


Specific Operations 10-83 


Duct and mobile hood 


maich here PAs 


Die hoist -\. 


—~ 


ee 


{ 
~- Hood travels on die 
hoist crane 


Q = 300WL 
Minimum duct velocity = 2000 fpm 
Hhe= 0.25 VPy 


ERICAN CONFERENCE 
: OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


AMI 


MOBILE) HOOD 
DIE’ CASTING 


“TaD DOT — : SGURE OT, Ash _ a7 


10-84 Industrial Ventilation 


10.60 MIXING 


Mixing operations combine a large variety of materials, 
usually without significant chemical reactions. This section 
includes categories of mixing operations. 


10.60.1 Mixing and Mulling: Mixers and mullers require 
exhaust ventilation to provide a minimum velocity of 150 fpm 
through all openings. Additional ventilation may be required 
when flammable solvents are used. The dilution ventilation 
rates should maintain concentrations within the muller below 
25% of the Lower Explosive Limit (LEL). Some codes or 
standards may require ventilation rates which ensure the 
concentration of flammable vapor is maintained below 20% 
of the LEL. 


10.60.2 Roll Mixing: The machinery shown in this subsec- 
tion is used to mix and blend quantities of viscid materials, 
such as rubber and plastic, with additives that are dry powers 
or liquids. Emissions of gases and vapors may evolve due to 


chemical reactions or may be caused by the elevated tempera- 
ture of the mixed materials. Particulate emissions can occur 
during additions as well as during mechanical blending. 


The roller mill ventilation shown in VS-60-12 encloses the 
roller mill to the maximum extent possible except for a front 
opening of sufficient height and width to permit operator 
access and material entry/removal. An air curtain directed 
upward towards the top of the enclosure provides a barrier to 
contaminant escape but still permits operator access and 
material entry/removal. The design values provided are criti- 
cal for proper operation. 


REFERENCES 


10.60.1 Hamp], V.; Johnston, O.E.; Murdock, D.M.: Appli- 
cation of an Air Curtain Exhaust System at a Milling 
Process. American Industrial Hygiene Association 
Journal, 49(4):167-175 (1988). 


Specific Operations 10-85 


——To prevent condensation, insulation, 
strip heaters or dilution fitting may 
be necessary. ee 


le Enclosing 
hood 


Skip hoist hood 
Q = 250 LW 


Baffle 
\ ! 


|Skip hoist hood Snares 
fh, =1.78 VP, + 0.25 VP, 


Opening for — | 
skip loading 


= 150 cfm/ft through all openings but not less than: 


[Muller « diam. _feet| ‘Exhaust, — 


Minimum duct velocity = 4000 fpm 
he = 0.25 VPy 

Other types of mixers: enclose as much as possible and provide 150 cfm/ft 
of remaining openings 
When flammable solvents are used in mixer, calculate minimum exhaust rate 
“for dilution to 25% of the LEL. See Chapter 2 


For air cooled mullers see VS-60-02 


AMERICAN CO NPR RED NCE | ete ny tases eatind- Oc, ; 
| x AALN. ) d bya 4 a, SJ. 

De Ove MIXER AND MULLER HOOD 
IND USTRIAL, HYGIENIS TS B rm 2 = = = —————— 2m FSIS a7 as — — OT ; 


10-86 Industrial Ventilation 


—To prevent condensation, ; ~~ Low~velocity ; 
Loading-— insulation, strip heaters duct used with 
hopper —_ or dilution fitting may : ] cooling type 
| ; be necessary. Soeeesaan ae i muller. 
sh 


zB =) renee paral 


booth 


! 
Bond hopper——~) 


i 
| 
i 
i 
i 


Cooling fan 
™" blow-through 
arrangement 


Muller Enclosing hood 


| Muller 


Ue ets aie cs 


Minimum exhaust rate, cfm 


Location R : Blow—though | Draw—though 
cae Sree cooling cooling 


| Batch hopper 
Bond hopper 

' Muller: 

4’ diameter 
6’ diameter 
7’ diameter 
8’ diarneter 
10° diameter 


fpm 

he = 0.25 VP 
Batch hopper requires separate exhaust with blow-through cooling. With other fan 
arrangement (muller under suction), separate exhaust may not be required. (if skip 
hoist is used, see VS-60-01) 
Maintain 150 fpm velocity through all openings in muller hood. Exhaust flow rates 
shown are the minimum for control. 
Cooling mullers do not require additional exhaust if maintained in dust tight 
condition. Blow-through fan must be off during loading. If muller is not dust tight, 
exhaust as in note 2 plus cooling air flow rate. 
When flammable solvents are used in mixer, calculate minimum exhaust flow rate 


for dilution to 25% of the LEL. See Chapter 2. 


AMERICAN CONFERENCE J AIR COOLED 
OF GOVERNMENTAL MIXER AND MULLER 


Specific Operations 10-87 


‘a 

| 

t 

Charging hood Pee 
; | 


. 


To suit method of -~ 
feed to mixer. 


Platform me 


gXes. accwaaeert oeaoneaion 
- 6 Diam. duct 


Te, 
(Qu 800 cfm* 


~ Trunnion exhaust 


= 200-300 cfm /tt open face area. 
500 cfm/ft of belt width if belt feeder used. 
velocity = 3500 fpm minimum. 

= 0.25 VR; at hood 
1.0 VFR; at trunnion 


* Consult mixer manufacturer for specific recommendations. 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 


B INDUSTRIAL HYGIENISTS PETES OT 


BANBURY MIXER 


10-88 Industrial Ventilation 


Side baffles 
desirable —--~ 


fie Better location of 
brake bar 


Q = 125 cfm/ft open face area 
Minimum duct veiocity = 2000 fpm 


he= 0.25 VP, 


NOTE: Both sides may be ooen 


AMERICAN CONFERENCE | ppponp parennpp pair] 
INDUSTRIAL HYGIENISTS | 


Een 
2 


Vs 


Specific Operations 10-89 


Material feed 7 


Operator opening - 


2? 


See note 4-—~\_ 


FRONT VIEW 


0.145 \/L ft (Manifold diameter) 
29.6 L cfm (Supply flow) 
r = 320 \V/X L cfm (Exhaust flow) 


a x 0.25” Holes (diam.) 
0.75" Apart (center to center) 


Manifold pressure = +1.2 “wg 


Notes: 1. All dimensions in feet, except as otherwise noted. 
2. For access openings other than operator opening, 
increase Q, by 100 cfm/ ft of opening area. 
lf operator opening is required on both sides of mill, 
total Q, will be sum of Q, for both sides. 
. X not to exceed 6 ff. 


| Reference 10.60.1 


AMERICAN CONFERENCE SAE TR APTRD SORRY RTA 
: i aie Ld dads | FLIN Aha LI j wa 
OF GOVERNMENTAL ROLLER MILL VENTILATION 


| INDUSTRIAL HYGIENISTS Kee 


10-90 Industrial Ventilation 


10.65 MOVABLE EXHAUST HOODS 


Movable exhaust hoods provide contro] for moving con- 
taminant sources. In general, movable hoods are associated 
with flexible exhaust ducts; traveling exhaust hoods, swivel, 
slip, or telescoping joints in duct sections; or systems which 
separate the hood from the duct for access to the process. 


Flexible exhaust duct is possibly the most common way of 
providing a movable exhaust hood. A section of flexible duct 
connects to a relatively small exhaust hood. The duct section 
and hood may be supported by a counter-weighted or spring- 
loaded, hinged arm that allows the positioning of the exhaust 
hood near the source of contaminant generation. This type of 
device is known as "snorkel," "elephant trunk,” or "flex-arm" 
exhaust. Illustrations of this type of exhaust in Chapter 10 
include Welding Exhaust (VS-65-01) and Granite Cutting and 
Finishing (VS-65-02). Flexible exhaust duct use is also illus- 
trated in Barrel Filling (VS-15-01), Metal Spraying (VS-90- 
30) and Service Garage Ventilation (VS-85-01 and 
VS-85-02) and low volume/high velocity systems (Section 
10.40). Frictional resistance can be very high in the flexible 
duct section as can the negative pressure or suction. Materials 
used in the construction of the duct may be metal or non-metal 
and the losses vary over a wide range depending on the type 
and use. The application data provided by the manufacturer 
must be included in the design development of the system. 
When used, flexible duct should be non-collapsible with 
minimal length to reduce undesirable bends which will result 
in excessive static pressure losses. 


Traveling exhaust hoods may be used for a variety of 
operations where the contaminant source moves from one 
point to another. This type is more suited to heavy-duty 
requirements than the flexible exhaust duct. Examples of 


these operations include flame and plasma cutting, foundry 
pouring, heavy abrasive cutting, and similar operations. An 
illustration of a traveling exhaust hood is the Hawley Trav-L- 
Vent (VS-65-03).(10-.6.) 


Telescoping or slip joints are duct sections that overlap, 
slide, or rotate to allow a section of one exhaust duct to slide 
into or rotate around another section of duct. This arrangement 
allows an exhaust hood to be moved from one position to 
another without disconnecting the duct or hood. Illustrations 
of the slip joint include the Core Grinder (VS-80-13) and 
Granite Cutting and Finishing (VS-65-02). This joint or duct 
section may include a swivel feature to permit rotation of the 
hood away from the process equipment being exhausted and 
may be used in the horizontal or vertical plane. Guide rails 
may be required for the horizontal application while pulleys 
and counter-weights may be required for vertical applications. 


Separating exhaust duct sections is another method of 
providing a movable exhaust hood. This concept requires that 
the exhaust duct separate at or near the exhaust hood when 
the hood is moved for access to the process equipment or the 
process equipment moves. Illustrations of this method include 
hoods for Top Electrode Melting Furnace (VS-55-03), Core 
Making Machine—Small Roll-Over Type (VS-20-11), and 
Mobile Hood, Die Casting (VS-55-21). Alignment of the 
exhaust duct when the hood is in place is critical, and any 
opening at this point should be included in the total exhaust 
calculations. Also, during the separation period, little or no 
exhaust control will be available at the contaminant source. 


REFERENCES 


10.65.1 Vulcan Engineering Co.: Hawley Trav-L-Vent 
Equipment Specifications and Layout. Helena, AL. 


Specific Operations 10-91 


Overhead 
res Ae 


Galvanized hood- 


FLEXIBLE EXHAUST CONNECTIONS 


Support 
Swivel base and angle ™~ 
iron duct support arm — 


Telescoping flex 
duct support ~ 


- Swivel joints ————-—~~~ 


AMERICAN CONFERENCE | |) 
OF GOVERNMENTAL MOVEABLE EXHAUST HOODS 


a 


g INDUSTRIAL HYGIENISTS aE —7S=o7 ROE SOS OT 


10-92 Industrial Ventilation 


4” |.D. flex. duct (rubber) or metal 
duct with telescope joints 


Revolving 
head 


bs Branch duct 


' Chip trap 
- x 8 opening with 
) 3” metal flange 


Q = 400 cfm minimum, tool 10° max distance from hood 
Minimum duct velocity = 4000 fpm 


_-— Flexible duct to 
branch duct 


- Attach to machine 


Abrasive blasting to be done in a room 
or cabinet; 500 fom at all openings. See 
"Abrasive Blasting , VS-80-01 


SURFACE MACHINE HOODS 


Hood cfm Branch 


Baby surfacer 400 4 
Medium surfacer 600 5” 


he= 1.0 VPy (at point A) 


AMERICAN CONFERENCE | GRANITE CUTTING 
OF GOVERNMENTAL | AND FINISHING 


J INDUSTRIAL HYGIENISTS | TE ENTER 


Specific Operations 10-93 


Exhaust 


SS carriage Z 
travel 
. <4 J 


Exhaust 
flow carriage \ 
* travel | 


| Exhaust : 
i < ee air : 


flow 


“—~ Exhaust 
hood 


Typical Design Ranges 


2500 — 10,000 cfm 
5500 — 4000 fom 
6” 


wg approx. total 


SP, = calculate separately 


Note: This is a patented system with many variations. 
Consult manufacturer for applications. 


: AMERICAN CONFERENCE 
: OF GOVERNMENTAL 


INDUSTRIAL HYGIENISTS 


HAWLEY TRAV-—L~-VENT 


PERSPECTIVE LAYOUT : 


10-94 Industrial Ventilation 


10.70 OPEN SURFACE TANKS 


Ventilation rates for plating, cleaning, and other open sur- 
face tank operations will depend on a number of parameters 
which include materials, tank configuration and location, and 
type of ventilation system. This section describes four 
hood/ventilation types: enclosing and canopy hoods, lateral 
exhaust, and push-pull. 


Enclosing hoods usually consist of a lateral hood with one 
end panel (two sides open) or panels at both tank ends (one 
side open). This hood configuration can provide increased 
efficiency by reducing the effects of cross-drafts and by 
directing more of the hood air flow over the tank open surface. 


Canopy hoods may be open on four sides (free standing) or 
on three sides (such as against a wall). Control is achieved by 
air flow into the hood. It is, however, difficult in many cases 
to achieve sufficient control velocity without excessive air 
flow rates. Canopy hoods should not be used with highly toxic 
materials, in locations where high cross-drafts are unavoid- 
able, or where the worker must bend over the tank. 


Lateral exhaust consists of a slot hood which controls 
emissions by pulling air across the tank. A single slot may be 
used on one side of the tank where the tank width is 36 inches 
or less. For widths greater than 36 inches and where the 
process configuration will allow, two slot hoods on opposite 
sides of the tank or a slot hood along the tank centerline may 
be used. A single slot may be used up to a tank width of 48 
inches but only if the material hazard class is low and if 
cross-drafts are not present.(see Section 10.70.1) 


The air flow required will be that necessary to achieve a 
minimum control velocity determined by the hazard class of 
the material used for operation and the particular tank/venti- 
lation system configuration. The procedure for determining 
the class and minimum control velocity for the three preced- 
ing hood types is provided in Tables 10.70.1—10.70.7 and the 
accompanying text. Exhaust flow for a canopy hood is deter- 
mined from VS-99-03 and for a booth hood from Figure 3-11 
where W is the total opening width. The exhaust flow for a 
lateral hood is determined from Table 10.70.4. 


Air and/or mechanical agitation of the tank solution may 
be used as an aid to the plating or cleaning process. Mechani- 
cal agitation creates a rolling motion and usually will not 
affect tank emissions. However, air agitation creates a boil- 
ing-like condition and may significantly increase tank emis- 
sions, thus creating need for increased exhaust flow to provide 
effective control. 


Push-pull ventilation consists of a push jet located on one 
side of a tank with a lateral exhaust hood on the other 
side.C071-10.703) Tank emissions are controlled by the jet 
formed over the tank surface. The jet captures the emissions 
and carries them into the hood. As the jet velocity, at all 
locations across the tank, is higher than the maximum control 
velocities specified for canopy, enclosing, or lateral exhaust 


hoods (Table 10.70.3), the push-pull exhaust flow is deter- 
mined on the basis of that necessary to capture the jet flow 
and is independent of the hazard classification. Push-pull 
design criteria are provided in VS-70-10, -11 and -12. 


10.70.1 Tank Design Considerations: 


1. Duct velocity = any desired velocity (see Chapter 3). 


2. Entry loss = 1.78 slot VP plus duct entry loss for slot 
hoods. For canopy or enclosure hoods, entry loss = 
duct entry loss. 


3. Maximum slot hood plenum velocity = 1/2 slot veloc- 
ity (see Chapter 3). 


4. Slot velocity = 2000 fpm unless distribution is pro- 
vided by well-designed, tapered takeoff. 


5. Provide ample area at the small end of the plenum. 


6. If L = 6' or greater, multiple takeoffs are desirable. If 
L = 10' or greater, multiple takeoffs are necessary. 


~— 


. Tank width (W) means the effective width over which 
the hood must pull air to operate (e.g., where the hood 
face is set back from the edge of the tank, this setback 
must be added in measuring tank width). 


If W = 20", slot on one side is suitable. 
If W = 20 — 36", slots on both sides are desirable. 


If W = 36 — 48", use slots on both sides or along tank 
centerline or use push-pull. A single slot along one side 
should not be used unless all other conditions are 
optimum. 


If W = 48" or greater, local exhaust usually is not 
practical. Consider using push-pull. 


Enclosure can be used for any width tank if process 
will permit. 


It is not practicable to ventilate across the long dimen- 
sion of a tank whose ratio W/L exceeds 2.0. It is 
undesirable to do so when W/L exceeds 1.0 


8. Liquid level should be 6" to 8" below top of tank with 
parts immersed. 


9. Lateral hood types A, C, D and E (VS-70-01 and -02) 
are preferred. Plenum acts as baffle to room air cur- 
rents. 


10. Provide removable covers on tank if possible. 


11. Provide duct with cleanouts, drains and corrosion-re- 
sistant coating if necessary. Use flexible connection at 
fan inlet. 


12. Install baffles to reduce cross-drafts. A baffle is a 
vertical plate the same length as the tank and with the 
top of the plate as high as the tank is wide. If the exhaust 
hood is on the side of the tank against a building wall 
or close to it, it is perfectly baffled. 


13. Replacement air to the tank area must be supplied 
evenly and directed toward the tank from above or in 
front of the tank so that cross-drafts do not occur. 


Flow Rate Calculation for Good Conditions: (No cross- 
crafts, adequate and well-distributed replacement air.) 


1. Establish process class by determining hazard poten- 
tial from Tables 10.70.1 and 10.70.2; information from 
Threshold Limit Values, Solvent Flash Point, Solvent 
Drying Time Tables in Appendices A and B and 
Tables 10.70.5-10.70.8. 


2. Process class can also be established directly from 
Tables 10.70.5-10.70.8 if process parameters are 
known. 


3. From Table 10.70.3, choose minimum control velocity 
according to hazard potential; evolution rate (process 
class); and hood design (see Table 10.70.5 for typical 
processes). 


4. From Table 10.70.4, select the cfm/ft? for tank dimen- 
sions and tank location. 


5. Multiply tank area by value obtained from Table 
10.70.4 to calculate required air volume. 


EXAMPLE 
Given: Chrome Plating Tank 6' x 2.5" 


High production decorative chrome 
Free standing in room 
No cross-drafts 


a. Tank Hood. See VS-70-01. Use hood "A" long 6' 
side. Hood acts as baffle 


b. Component — Chromic Acid (Chromium, metal; 
water-soluble CrVIJ) 
Hazard potential: A (from Table 10.70.1; from Ap- 
pendix A: TLV = 0.05 mg/m}; from Appendix A: 
Flash point = Negligible) 
Rate of Evolution: 1 from Table 10.70.2; from Table 
10.70.6: Gassing rate = high) 
Class: A-1 
Control Velocity = 150 fpm (from Table 10.70.3) 


Minimum Exhaust Rate = 225 cfm/ft? (from Table 
10.70.4; Baffled tank, W/L = 0.42) 


Minimum Exhaust Flow Rate = 225 x 15 =3375 cfm 
c. Hood Design 

Design slot velocity = 2000 fpm 

Slot area = Q/V = 3375 cfm/2000 fpm = 1.69 ft? 

Slot Width = A/L = 1.69 ft?/6 ft = 0.281' = 3.375" 


Example (continued) 


Plenum depth = (2)(slot width) = (2)(3.375) = 6.75" 


Specific Operations 10-95 


TABLE 10.70.1. Determination of Hazard Potential 


HYGIENIC STANDARDS 
Hazard Gas and Vapor Mist Flash Point 
Potential (see Appendix A) (see Appendix A) (see Appendix B) 
A 0-10 ppm 0-0.1 mg/m — 
B 11-100 ppm ~=—0.11-1.0 mg/m? ——- Under 100 F 
C 101-500 ppm ~—-1.1-10 mg/m? 100-200 F 
D Over 500 ppm —_ Over 10 mg/m’ Over 200 F 


TABLE 10.70.2. Determination of Rate of Gas, Vapor, 
or Mist Evolution 


Degrees Relative Evaporation* 
Liquid Below Boiling (Time for 100% 
Rate Temperature (F) Point (F) Evaporation) Gassing** 
{ Over 200 0-20 Fast (0-3 hours) High 
2 150-200 21-50 = Medium (3-12 hours) Medium 
3 94-149 51-100 Slow (12-50 hours) Low 
4 Under 94 Over 100 Nil (Over 50 hours) Nil 


*Dry Time Relation (see Appendix B). Below 5 — Fast; 5~15 — Medium; 
15-75 — Slow; 70-over — Nil. 


™Rate of gassing depends on rate of chemical or electrochemical action and 
therefore depends on the material treated and the solution used in the tank and 
tends to increase with 1) amount of work in the tank at any one time, 2) strength of 
the solution in the tank, 3) temperature of the solution in the tank, and 4) current 
density applied to the work in electrochemical tanks. 


Duct area = Q/V = 3375 cfm/2500 fpm = 1.35 ft? 
Use 16" duct, area = 1.396 ft? 
Final duct velocity = Q/A = 3375/1.396 = 2420 fpm 
Hood SP= Entry loss + Acceleration 
1.78 VP, + 0.25 VPq + 1.0 VP (see 


i 


Chapter 3) 
= (1.78 x 0.25") + (0.25 x 0.37") + 0.37" 
= 0.45 + 0.09 + 0.37 
Hood SP= 0.91" 
REFERENCES 


10.70.1 Huebener, D.J.; Hughes, R.T.: Development of 
Push-Pull Ventilation. American Industrial Hygiene 
Assoc. Journal, Vol. 46:262—267 (1985). 


10.70.2 Hughes, R.T.: Design Criteria for Plating Tank 
Push-Pull Ventilation. In: Ventilation ‘85: Proceed- 
ings of the First International Symposium on Venti- 
lation for Contaminant Control. Elsiever Press, 
Amsterdam, the Netherlands (1986). 


10.70.3 Sciola, V.: Private Communication, Hamilton 
Standard. 


10-96 Industrial Ventilation 


TABLE 10.70.3. Minimum Control Velocity (FPM) for Undisturbed Locations 


Canopy Hoods 
Class (see Tables Enclosing Hood Lateral Exhaust (see Figure 3-8 & VS-99-03) 
10.70.1 & 10.70.2) One Open Side Two Open Sides (see VA-70-01 & 70-02) (Note 1) Three Open Sides Four Open Sides 
A-1 and A-2 (Note 2) 100 150 150 Do not use Do not use 
A-3 (Note 2), B-1, B-2, and C-1 75 100 100 125 175 
B-3, C-2, and D-1 (Note 3) 65 90 75 100 150 
A-4 (Note 2) C-3, and 50 75 50 75 125 


D-2 (Note 3) 
B-4, C-4, D-3 (Note 3), and D-4 — Adequate General Room Ventilation Required (see Chapter 2). 


Notes: 1. Use aspect ratio to determine air volume; see Table 10.70.4 for computation. 
2. Do not use canopy hood for Hazard Potential A processes. 
3, Where complete control of hot water is desired, design as next highest class. 


TABLE 10.70.4. Minimum Rate, cfm/ft? of Tank Area for Lateral Exhaust 


cfm/ft? to maintain required minimum control velocities at following 
tank width (* 


— | ratios 
tank length \L } alii 


Required Minimum 
Control Velocity, fpm 


(from Table 10.70.3) 0.0-0.09 0.1-0.24 0.25-0.49 0.5-0.99 1.0-2.0 (Note 2) 
Hood against wall or flanged (see Note 1 below and Section 10.70.1. Note 12). See VS-70-01 A and VS-70-02 D and E. 
50 50 60 rhs) 90 100 
75 75 90 110 130 150 
100 100 125 150 175 200 
150 150 190 225 [250] Note 3 [250] Note 3 
Hood on free standing tank (see Note 1). See VS-70-01 B and VS-70-02 F. 
50 75 90 100 110 125 
75 110 130 150 170 190 
100 150 175 200 225 250 
150 225 [250] Note 3 [250] Note 3 [250] Note 3 [250] Note 3 


Notes: 1, Use W/2 as tank width in computing W/L ratio for hood along centerline or two parallel sides of tank. See VS-70-01 B and C and VS-70-02 F. 
2. See Section 10.70.1, Notes 6 and 7. 
3. While bracketed values may not produce 150 fpm control velocity at all aspect ratios, the 250 cfmvft’ is considered adequate for control. 


Specific Operations 10-97 


TABLE 10.70.5. Typical Processes Minimum Control Velocity (fpm) for Undisturbed Locations 


Lateral Exhaust 
Control Velocity 


Contaminant (See VS-70-01 & Collector 
Operation Contaminant Hazard Evolution VS-70-02) Recommended 
Anodizing Aluminum Chromic-Sulfuric Acids A 1 150 X 
Aluminum Bright Dip Nitric + Sulfuric Acids A 1 150 x 
Nitric + Phosphoric Acids A | 150 X 

Plating — Chromium Chromic Acid A 1 150 X 

Copper Strike Cyanide Mist C 2 75 X 
Metal Cleaning (Boiling) Alkaline Mlst Cc 1 100 X 
Hot Water (if vent desired) 

Not Boiling Water Vapor D 2 50* 

Boiling D 1 75* 
Stripping — Copper Alkaline-Cyanide Mists C 2 75 X 

Nickel Nitrogen Oxide Gases A 1 150 X 
Pickling — Steel Hydrochloric Acid A 2 150 X 

Sulfuric Acid B 1 100 X 

Salt Solution 

Bonderizing & Parkerizing) Water Vapor D 2 50* 

Not Boiling Water Vapor D 2 50* 

Boiling D { 75* 
Salt Baths (Molten) Alkaline Mist C 1 100 X 


*Where complete control of water vapor is desired, design as next highest class. 


TABLE 10.70.6. 


Airborne Contaminants Released by Metallic Surfaced Treatment, Etching, Pickling, Acid Dipping and Metal Cleaning Operations 


Component of Bath Which May be Physical and Chemical Nature of Usual Temp. 
Process Type Notes Released to Atmosphere (13) Major Atmospheric Contaminant Class (12) Range F 
Surface Treatment Anodizing Aluminum Chromic-Sulfuric Acids Chromic Acid Mist A-1 95 
Anodizing Aluminum Sulfuric Acid Sulfuric Acid Mist B-4 60-80 
Black Magic Conc. Sol. Alkaline Oxidizing Agents Alkaline Mist, Steam C-1 260-350 
Bonderizing 1 Boiling Water Steam D-2,1 (14, 15) 140-212 
Chemical Coloring None None D-4 70-90 
Descaling 2 Nitric-Sulfuric, Hydrofluoric Acids Acid Mist, Hydrogen Fluoride Gas, Steam B-2,1 (15) 70-150 
Ebonol Conc. Sol. Alkaline Oxidizing Agents Alkaline Mist, Steam C-1 260-350 
Galvanic-Anodize 3 Ammonium Hydroxide Ammonia Gas, Steam B-3 140 
Hard-Coating Aluminum Chromic-Sulfuric Acids Chromic Acid Mist A-1 120-180 
Hard-Coating Aluminum Sulfuric Acid Sulfuric Acid Mist B-1 120-180 
Jetal Conc. Sol. Alkaline Oxidizing Agents Alkaline Mist, Steam C-1 260-350 
Magcote 4 Sodium Hydroxide Alkaline Mist, Steam C-3,2 (15) 105-212 
Magnesium Pre-Dye Dip Ammonium Hydroxide-Ammonium Acetate Ammonia Gas, Steam B-3 90-180 
Parkerizing 1 Boiling Water Steam D-2,1 (14,15) 140-212 
Zincete Immersion 5 None None D-4 70-90 
Etching Aluminum Sodium Hydroxide-Soda Ash-Trisodium Alkaline Mist, Steam C-1 160-180 
Phosphate 
Copper 6 Hydrochloric Acid Hydrogen Chloride Gas A-2 70-90 
Copper 7 None None D-4 70 
Pickling Aluminum Nitric Acid Nitrogen Oxide Gases A-2 70-90 
Aluminum Chromic, Sulfuric Acids Acid Mists A-3 140 
Aluminum Sodium Hydroxide Alkaline Mist C-1 140 
Cast Iron Hydrofluoric-Nitric Acids Hydrogen Fluoride-Nitrogen Oxide Gases A-2,1 (15) 70-90 
Copper Sulfuric Acid Acid Mist, Steam B-3,2 (15) 125-175 
Copper 8 None None D-4 70-175 
Duralumin Sodium Flouride, Sulfuric Acid Hydrogen Fluoride Gas, Acid Mist A-3 70 
Inconel Nitric, Hydrofluoric Acids Nitrogen Oxide, HF Gases, Steam A-1 150-165 
Inconel Sulfuric Acid Sulfuric Acid Mist, Steam B-2 160-180 
Iron and Steel Hydrochloric Acid Hydrogen Chloride Gas A-2 70 
Iron and Steel Sulfuric Acid Sulfuric Acid Mist, Steam B-1 70-175 
Magnesium Chromic-Suifuric, Nitric Acids Nitrogen Oxide Gases, Acid Mist, Steam A-2 70-160 
Monel and Nickel Hydrochloric Acid Hydrogen Chloride Gas, Steam A-2 180 
Monel and Nickel Sulfuric Acid Sulfuric Acid Mist, Steam B-1 160-190 
Nickel Silver Sulfuric Acid Acid Mist, Steam B-3,2 (15) 70-140 
Silver Sodium Cyanide Cyanide Mist, Steam C-3 70-210 
Stainless Steel 9 Nitric, Hydrofluoric Acids Nitrogen Oxide, Hydrogen Fluoride Gases A-2 125-180 
Stainless Steel 9,10 Hydrochloric Acid Hydrogen Chloride Gas A-2 130-140 
Stainless Steel 9,10 Sulfuric Acid Sulfuric Acid Mist, Steam B-1 180 
Stainless Steel Immunization Nitric Acid Nitrogen Oxide Gases A-2 70-120 
Stainless Steel Passivation Nitric Acid Nitrogen Oxide Gases A-2 70-120 


86-01 


uonEpUa, [eLysnpuy 


TABLE 10.70.6. 


Airborne Contaminants Released by Metallic Surfaced Treatment, Etching, Pickling, Acid Dipping and Metal Cleaning Operations (con’t) 


Component of Bath Which May be Physical and Chemical Nature of Usual Temp. 
Process Type Notes Released to Atmosphere (13) Major Atmospheric Contaminant Class (12) Range F 
Acid Dipping Aluminum Bright Dip Phosphoric, Nitric Acids Nitrogen Oxide Gases A-1 200 
Aluminum Bright Dip Nitric, Sulfuric Acids Nitrogen Oxide Gases, Acid Mist A-2,1 (15) 70-90 
Cadmium Bright Dip None None D-4 70 
Copper Bright Dip Nitric, Sulfuric Acids Nitrogen Oxide Gases, Acid Mist A-2,1 (15) 70-90 
Copper Semi-Bright Dip Sulfuric Acid Acid Mist B-2 70 
Copper Alloys Bright Dip Nitric, Sulfuric Acids Nitrogen Oxide Gases, Acid Mist A-2,1 (15) 70-90 
Copper Matte Dip Nitric, Sulfuric Acids Nitrogen Oxide Gases, Acid Mist A-2,1 (15) 70-90 
Magnesium Dip Chromic Acid Acid Mist, Steam A-2 190-212 
Magnesium Dip Nitric, Sulfuric Acids Nitrogen Oxide Gases, Acid Mist A-2,1 (15) 70-90 
Monel Dip Nitric, Sulfuric Acids Nitrogen Oxide Gases, Acid Mist A-2,1 (15) 70-90 
Nickel and Nickel Alloys Dip Nitric, Sulfuric Acids Nitrogen Oxide Gases, Acid Mist A-2,1 (15) 70-90 
Silver Dip Nitric Acid Nitrogen Oxide Gases A-1 70-90 
Silver Dip Sulfuric Acid Sulfuric Acid Mist B-2 70-90 
Zinc and Zinc Alloys Dip Chromic, Hydrochloric Acids Hydrogen Chloride Gas (If HC! A-4,3 (15) 70-90 
attacks Zn) 
Metal Cleaning Alkaline Cleaning 14 Alkaline Sodium Salts Alkaline Mist, Steam C-2,1 (15) 160-210 
Degreasing Trichloroethylene-Perchloroethylene Trichloroethylene-Perchloroethylene B (16) 188-250 
Vapors 
Emulsion Cleaning Petroleum-Coal Tar Solvents Petroleum-Coal Tar Vapors B-3,2 (15) 70-140 
(17) 70-140 
Emulsion Cleaning Chlorinated Hydrocarbons Chlorinated Hydrocarbon Vapors (17) 70-140 


Notes: 1 Also Aluminum Seal, Magnesium Seal, Magnesium Dye Set, Dyeing Anodized 
Magnesium, Magnesium Alkaline Dichromate Soak, Coloring Anodized Aluminum. 


2 Stainless Steel Before Electropolishing. 
3 On Magnesium. 
4 Also Manodyz, Dow-12. 


5 On Aluminum. 


6 Dull Finish. 


7 Ferric Chloride Bath. 

8 Sodium Dichromate, Sulfuric Acid Bath and Ferrous Sulfate, Sulfuric Acid Bath. 

9 Scale Removal. 
10 Scale Loosening. 


11 Soak and Electrocleaning. 


12 Class as described in Table 10.70.3 based on hazard potential (Table 10.70.1) 


and rate of evolution (Table 10.70.2) for usual operating conditions. Higher temperatures, agitation 
or other conditions may result in a higher rate of evolution. 


13 Hydrogen gas also released by many of these operations. 


14 Rate where essentially complete control of steam is required. Otherwise, adequate dilution 


ventilation may be sufficient. 


15 The higher rate is associated with the higher value in the temperature range. 
16 For vapor degreasers, rate is determined by operating procedure. See VS-70-20. 
17 Class of operation is determined by nature of the hydrocarbon. Refer to Appendix A. 


suonesdg syreds 


66-01 


TABLE 10.70.7. Airborne Contaminants Released by Electropolishing, Electroplating and Electroless Plating Operations 


Process 


Component of Bath Which May be Physical and Chemical Nature of Usual Temp. 
Type Notes Released to Atmosphere (19) Major Atmospheric Contaminant Class (18) Range F 
Electropolishing Aluminum 1 Sulfuric, Hydrofiuoric Acids Acid Mist, Hydrogen Flouride Gas, Steam A-2 140-200 
Brass, Bronze 1 Phosphoric Acid Acid Mist B-3 68 
Copper 1 Phosphoric Acid Acid Mist B-3 68 
iron 1 Sulfuric, Hydrochloric, Perchloric Acids Acid Mist, Hydrogen Chloride Gas, Steam  A-2 68-175 
Monel 1 Sulfuric Acid Acid Mist, Steam B-2 86-160 
Nickel 1 Sulfuric Acid Acid Mist, Steam B-2 86-160 
Stainless Steel 1 Sulfuric, Hydrofluoric, Chromic Acids Acid Mist, Hydrogen Flouride Gas, Steam A-2,1 (20) 70-300 
Steel 1 Sulfuric, Hydrochloric, Perchloric Acids Acid Mist, Hydrogen Chloride Gas, Steam A-2 68-175 
Strike Solutions Copper Cyanide Salts Cyanide Mist C-2 70-90 
Silver Cyanide Salts Cyanide Mists C-2 70-90 
Wood's Nickel Nickel Chloride, Hydrochloric Acid Hydrogen Chloride Gas, Chloride Mist A-2 70-90 
Electroless Plating Copper Formaldehyde Formaldehyde Gas A-1 75 
Nickel 2 Ammonium Hydroxide Ammonia Gas B-1 190 
Electroplating Alkaline Platium Ammonium Phosphate, Ammonia Gas Ammonia Gas B-2 158-203 
Tin Sodium Stannate Tin Salt Mist, Steam C-3 140-170 
Zinc 3 None None D-4 170-180 
Electroplating Fluoborate Cadmium Fluoborate Salts Fluoborate Mist, Steam C-3,2 (20) 70-170 
Copper Copper Fluoborate Fluoborate Mist, Steam C-3,2 (20) 70-170 
Indium Fluoborate Salts Fluoborate Mist, Steam C-3,2 (20) 70-170 
Lead Lead Fluoborate-Fluoboric Acid Fluoborate Mist, Hydrogen Fluoride Gas A-3 70-90 
Lead-Tin Alloy Lead Fluoborate-Fluoboric Acid Fluoborate Mist C-3,2 (20) 70-100 
Nickel Nickel Fluoborate Fluoborate Mist C-3,2 (20) 100-170 
Tin Stannous Fluoborate, Fluoboric Acid Fluoborate Mist C-3,2 (20) 70-100 
Zinc Fluoborate Salts Fluoborate Mist, Steam C-3,2 (20) 70-170 
Electroplating Cyanide Brass, Bronze 4,5 Cyanide Salts, Ammonium Hydroxide Cyanide Mist, Ammonia Gas B-4,3 (20) 60-100 
Bright Zinc 5 Cyanide Salts, Sodium Hydroxide Cyanide, Akaline Mists C-3 70-120 
Cadmium 5 None None D-4 70-100 
Copper 5,6 None None D-4 70-160 
Copper 5,7 Cyanide Salts, Sodium Hydroxide Cyanide, Alkaline Mists, Steam C-2 110-160 
Indium 5 Cyanide Salts, Sodium Hydroxide Cyanide, Alkaline Mists C-3 70-120 
Silver 5 None None D-4 72-120 
Tin-Zinc Alloy 5 Cyanide Salts, Potassium Hydroxide Cyanide, Alkaline Mists, Steam C-3,2 (20) 120-140 
White Alloy 5,8 Cyanide Salts, Sodium Stannate Cyanide, Alkaline Mists C-3 120-150 
Zinc 5,9 Cyanide Saits, Sodium Hydroxide Cyanide, Alkaline Mists C-3,2 (7) 70-120 


OOT-OL 


uonepyue, [BLYsnpUy 


TABLE 10.70.7. Airborne Contaminants Released by Electropolishing, Electroplating and Electroless Plating Operations (con’t) 


Notes: 


Component of Bath Which May be Physical and Chemical Nature of Usual Temp. 
Process Type Notes Released to Atmosphere (19) Major Atmospheric Contaminant Class (18) Range F 
Electroplating Acid Chromium Chromic Acid Chromic Acid Mists A-1 90-140 
Copper 10 Copper Sulfate, Sulfuric Acid Sulfuric Acid Mist B-4,3 (20,21) 75-120 
Indium 12 None None D-4 70-120 
Indium 13,14 Sulfamic Acid, Sulfamate Salts Sulfamate Mist C-3 70-90 
Iron Chloride Salts, Hydrochloric Acid Hydrochloric Acid Mist, Steam A-2 190-210 
Iron 12 None None D-4 70-120 
Nickel 3 Ammonium Fluoride, Hydrofluoric Acid Hydrofluoric Acid Mist A-3 102 
Nickel and Black 12,15 None None C-4 (22) 70-150 
Nickel 
Nickel 9,12 Nickel Sulfate Nickel Sulfate Mist B-2 70-90 
Nickel 13,14 Nickel Sulfamate Sulfamate Mist C-3 75-160 
Palladium 15 None None D-4 70-120 
Rhodium 12,17 None None D-4 70-120 
Tin Tin Halide Halide Mist C-2 70-90 
Tin 12 None None D-4 70-120 
Zinc Zinc Chloride Zinc Chloride Mist B-3 75-120 
Zinc 12 None None D-4 70-120 
1 Arsine may be produced due to the presence of arsenic in the metal or polishing bath. 14 Air Agitated. 


2 Alkaline Bath. 
3 On Magnesium. 
4 Also Copper-Cadmium Bronze. 


5 HCN gas may be evolved due to the acidic action of COz in the air at the surface of the bath. 


6 Conventional Cyanide Bath. 
7 Except Conventional Cyanide Bath. 


8 Albaloy, Spekwhite, Bonwhite (Alloys of Copper, Tin, Zinc). 


9 Using Insoluble Anodes. 
10 Over 90 F. 
11 Mild Organic Acid Bath. 
12 Sulfate Bath. 
13 Sulfamate Bath. 


15 Chloride Bath. 
16 Nitrite Bath. 
17 Phosphate Bath. 


18 Class as described in Table 10.70.3 based on hazard potential (Table 
10.70.1) and rate of evolution (Table 10.70.2) for usual operating conditions. 
Higher temperatures, agitation, high current density or other conditions may result 


in a higher rate of evolution. 


19 Hydrogen gas also released by many of these operations. 
20 The higher rate is associated with the higher value in the temperature range. 


21 Baths operated at a temperature of over 140 F with a current density of over 45 
amps/ft* and with air agitation will have a higher rate of evolution. 


22 Local exhaust ventilation may be desired to contro! steam and water vapor. 


suone19dg aypeads 


10r-01 


TABLE 10.70.8. Airborne Contaminants Released by Stripping Operations 


Base Metal Component of Batch Which May be Physical and Chemical Nature of Major Usual Temp. 
Coating to be Stripped (Footnote) Released to Atmosphere (f) Atmospheric Contaminant Class (e) Range F 
Anodized Coatings 1,7 Chromic Acid Acid Mist, Steam A-2 120-200 
Black Oxide Coatings 14 Hydrochloric Acid Hydrogen Chloride Gas A-3,2  (q) 70-125 
Brass and Bronze 8.14 (a) Sodium Hydroxide, Sodium Cyanide Alkaline, Cyanide Mists C-3,2 (g) 70-90 
Cadmium 8,14 (a) Sodium Hydroxide, Sodium Cyanide Alkaline, Cyanide Mists C-3,2  (g) 70-80 
2,4,14 Hydrochloric Acid Acid Mist, Hydrogen Chloride Gas A-3,2  (g) 70-90 
Chromium 7,8,14 (a) Sodium Hydroxide Alkaline Mist, Steam C-3 70-150 
2,4,8,14 Hydrochloric Acid Hydrogen Chloride Gas A-2 70-125 
2,4,8,18 (a) Sulfuric Acid Acid Mist B-2 70-90 
Copper 8,14 Sodium Hydroxide, Sodium Cyanide Alkaline, Cyanide Mists C-3,2 (9) 70-90 
7,12,14 (b) None None D-4 70-90 
14 (a) Alkaline Cyanide Cyanide Mist C-3,2 (g) 70-160 
1 Nitric Acid Nitrogen Oxide Gases A-1 70-120 
18 (a) Sodium Hydroxide-Sodium Sulfide Alkaline Mist, Steam C-2 185-195 
Gold 4,5,6,8,9,14 (a) Sodium Hydroxide, Sodium Cyanide Alkaline, Cyanide Mists C-3,2 (g) 70-90 
4,5,18 (a) Sulfuric Acid Acid Mist B-3,2 (g) 70-100 
Lead 13 (c) Acetic Acid, Hydrogen Peroxide Oxygen Mist D-3 70-90 
14 (a),(c) Sodium Hydroxide Alakaline Mist, Steam C-3,2 (g) 70-140 
Nickel 2,4 Sulfuric, Nitric Acids Nitrogen Oxide Gases A-2,1  (g) 70-90 
2,4 (a) Hydrochloric Acid Hydrogen Chloride Gas A-3 70-90 
2,4,14 (a) Sulfuric Acid Acid Mist B-3 70-90 
7 Hydrofluoric Acid Hydrogen Fluoride Gas A-3,2  (g) 70-90 
14 Fuming Nitric Acid Nitrogen Oxide Gases A-1 70-90 
(a),(d) Hot Water Steam D-2 (h) 200 
1,18,19 (a) Sulfuric Acid Acid Mist, Steam B-3,2  (g) 70-150 
Phosphate Coatings 15 Chromic Acid Acid Mist, Steam A-3 166 
16 Ammonium Hydroxide Ammonia Gas B-3,2 (g) 70-90 
Rhodium 10 Sulfuric, Hydrochloric Acids Acid Mist, Hydrogen Chloride Gas A-3,2  (g) 70-100 
Silver 1 Nitric Acid Nitrogen Oxide Gases A-1 70-90 
2,14 Sulfuric, Nitric Acids Nitrogen Oxide Gases, Steam A-1 180 
8,14 (a) Sodium Hydroxide, Sodium Cyanide Alkaline, Cyanide Mists C-3 70-90 
17 (a) Sodium Cyanide Cyanide Mist C-3 70-90 
Tin 2,3,4 Ferric Chloride, Copper Sulfate Acetic Acid Acid Mist B-4,3  (g) 70-90 
(a) Sodium Hydroxide Alkaline Mist C-3 70-90 
2,14 Hydrochloric Acid Hydrogen Chloride Gas A-3,2 (g) 70-90 
14 (a) Sodium Hydroxide Alkaline Mist, Steam C-2 70-200 


COT-OL 


uOHRQUIA [eLsNpuy 


TABLE 10.70.8. Airborne Contaminants Released by Stripping Operations (con’t) 


Base Metal Component of Batch Which May be Physical and Chemical Nature of Major Usual Temp. 
Coating to be Stripped (Footnote) Released to Atmosphere (f) Atmospheric Contaminant Class (e) Range F 
Zinc 1 Nitric Acid Nitrogen Oxide Gases A-4 70-90 
8,14 Sodium Hydroxide, Sodium Alkaline, Cyanide Mists C-3 70-90 
Base Metal: 1. Aluminum 8. Nickel 14. Steel 
2. Brass 9. Nickel Alloys 15. Steel (Manganese Type Coatings) 
3. Bronze 10. Nickel Plated Brass 16. Steel (Zinc Type Coatings) 
4. Copper 11. Nickel Silver 17, White Metal 
5. Copper Alloys 12. Non-Ferrous Metals 18. Zinc 
6. Ferrous Metals 13. Silver 19. Zinc Base Die Castings 
7. Magnesium 
Notes: (a) Electrolytic Process. (f} Hydrogen gas also released by some of these operations. 
(b) Refers only to steel (14) when Chromic, Sulfuric Acids Bath is used. (g) The higher rate is associated with the higher value in the temperature range. 
(c) Also Lead Alloys. (h) Rate where essentially complete control of steam is required. Otherwise, adequate dilution 
(d) Sodium Nitrate Bath. ventilation may be sufficient. 
(e) Class as described in Table 10.70. 3 based on hazard potential (Table10.70.1) and rate of evolution 


(Table 10.70.2) for usual operating conditions. Higher temperatures, agitation or other conditions 
may result in a higher rate of evolution. 


suonessdg ayiaads 


cOT-O1 


10-104 Industrial Ventilation 


A. UPWARD PLENUM 


Partial covers advisable if ——-—> | ; 
possible — on any type tank ,“ i be- 2S 


| eae = x aa 


Section A-A 


| AMERICAN CONFERENCE 
: OF GOVERNMENTAL 
P INDUSTRIAL HYGIENISTS 


OPEN SURFACE TANKS 


fare 7-90 [eure VS—7O0-O1 | 1 


Specific Operations 


after removal from 
tank. —--—-—- 


7 
Work gives off vapors 


| 


| To suit work 


Slot > 


Extend over tank as far 
ds, possible =. 


E. LATERAL 


" ! 
: Max. plenum velocity = 1/2 slot velocity 7 Ss 


F. END TAKE-OFF 


aa eee radius desirable if space permits 


Slot velocity 2000 fpm ~ 
Sloped plenum desirable 


AMERICAN CONFERENCE | 
f OF GOVERNMENTAL — | 
| INDUSTRIAL HYGIENISTS 


OPEN SURFACE TANKS 


10-105 


10-106 Industrial Ventilation 


Liquid Temperature (t) 


= 
< a 


Push nozzle manifold (1) -Circular, rectangular or square. Manifold cross-sectional : 
area should be at least 2.5 times the total nozzle flow area.| 


Push nozzle angle (2) — O° to 20° down. 


Nozzle openings @)- 1/8" to 1/4” slot or 5/32” to 1/4” dia. holes with 3 to 8 
dia. spacing. Outer holes or slot ends(4)must be 1/2” to 
1" inside tank inner edges. 


Exhaust opening (5) — Size to achieve 2000 fpm slot velocity. Outer edges of 
opening(7) must extend to edge of tank including flanges. 
Liquid surface (6) — Tank freeboard must not exceed 8” with parts removed. 
Push nozzle supply Q; = 243 VA; 
where Q; = push nozzle supply, cfm/ft manifold length 


Aj = total nozzle opening per foot of manifold length 
Total push supply Q, = Q; x L cfm 
Exhaust flow Q-= 75 cfm/ft” tank surface area for t £150 F 
Q,= (0.4 T + 15) cfm/ft* tank surface area for t > 150 F. 


Tank surface area = L (length of tank) x W (width of tank) 


Design Procedure: Select nozzle opening within above limits and calculate push supply 


and exhaust air flow. See VS—70-11 and VS—70-12. 


: Reference 10.70.14, 


10.70.2, & 10.70.3 


AMERICAN CONFERENCE 
f OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


PUSH-PULL HOOD DESIGN | 
DATA FOR WIDTHS UP TO 10’§ 


Specific Operations 10-107 


In push-pull ventilation, a nozzle pushes a jet of air across the vessel surface into an exhaust 
hood. Effectiveness of a push jet is a function of its momentum which can be related to the 
product of the nozzle supply air flow (Q;) and the nozzle exit velocity (V;). For a jet used for plating 
tanks or other open surface vessels, a push supply flow can be determined from: 


Q, = 243,/A; 


where: Qj = push nozzle supply, cfm per foot of push nozzle plenum length 
Aj = nozzle exit area, ft2/per foot of push nozzle plenum length 


Using this approach, a push nozzle design is first selected and the nozzle exit area (Aj) 
determined. 


The push nozzle manifold may be round, rectangular or square in cross-section. The push nozzle 
may be a 1/8" to 1/4" horizontal slot or 5/32" to 1/4" diameter drilled holes on 3 to 8 diameter 
spacing. 

It is important that the air flow from the nozzle be evenly distributed along the length of the supply 
plenum. To achieve this, the total nozzle exit area should not exceed 40% of the plenum 
cross-sectional area. Multiple supply plenum inlets should be used where practical. 


The push nozzle manifold should be located as near the vessel edge as possible to minimize 
the height above the liquid surface. The manifold should be adjustable to optimize the push jet 
angle. The manifold axis can be angled down a maximum of 200 to permit the jet to clear 
obstructions and to maintain the jet at the vessel surface. It is essential any opening between 
the manifold and tank be sealed. 


An exhaust flow of 75 cfm/ft2 of vessel surface area should be used for tank liquid temperatures 
(t) of 150 F or lower. For tank liquid temperatures greater than 150 F use an exhaust flow of (0.4t 
+ 15) cfm/ft2. These flow rates are independent of the “class” used in determining exhaust flow 
for side draft hoods. “Control velocity” is achieved by the push jet blowing over the tank and will 
be considerably higher than that which can be achieved by a side draft hood. The purpose of 
the exhaust hood is to capture and remove the jet—not to provide capture velocity. A flanged 
hood design is to be used wherever practical. The exhaust hood should be located at the vessel 
edge so as not to leave a gap between the hood and the vessel. 


Design and location of an open surface vessel encompasses a number of variables. In some 
cases vessel shape, room location, cross-drafts, etc., may create conditions requiring adjustment 
of the push and/or pull flow rates in order to achieve effective control. Cross-draft velocities over 
75 fpm, very wide vessels (eight feet or more), or very large or flat surface parts may require 
increased push and/or pull flows. To account for the effects of these variables, a flow adjustment 
of +20% should be designed into the push and +20% into the pull flow system. Wherever practical, 
construction and evaluation of a pilot system is recommended. Once designed and installed, 
push-pull systems can be initially evaluated by use of a visual tracer technique and appropriate 
flow adjustments can be made as required. 


The exhaust hood opening should be sized to assure even flow distribution across the opening. 
This can be achieved by sizing the slot for 2000 fpm slot velocity. 


AMERICAN CONFERENCE : DESIGN DATA 
OF GOVERNMENTAL PUSH= POLI « cdlOOD 


10-108 Industrial Ventilation 


"wg 


PUSH NOZZLE MANIFOLD STATIC PRESSURE — 


0.002 0.004 0.006 0.010 0.020 
NOZZLE FLOW AREA (A; ) a ft? /ft manifold length 


PUSH NOZZLE SUPPLY = 243 1A; cfm/ft of length for nozzles with 1/8 to 1/4 
inch wide slots or 5/32 to 1/4 inch diameter holes on 
3 to 8 diameter spacing. 


For holes Aj (ft* /ft)= 0.065 x hole diameter (in)/hote spacing (no. of diameters (in).) : 


slot width (in) 
12 


For slot A; (ft’ /ft)= (See VS—70-10) 


PUSH NOZZLE 
MANIFOLD PRESSURE 


| AMERICAN CONFERENCE | 


7 OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


Specific Operations 10-109 


No slot near ; 
take-off Inside radius 


desirable 


Maximum plenum 
velocity = 500 fom 


Cover when not in use 


Section A-A 


50 LW 
Slot velocity = 2000 fpm 
he= 1.78 VP5 + 0.25 VPy 


Duct velocity = 2000 minimum 


Also provide: 1. 


Separate flue for combustion products 
2. For cleaning operation, appropriate respiratory protection is necessary. 
3. For pit units, the pit should be mechanically ventilated. 

4. For further safeguards, see VS—501.1 


NOTE: Provide downdraft grille for pgrts that cannot be 
removed dry; Q = 50 cfm/ft* grille area. 


AMERICAN CONFERENCE 
: OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


SOLVENT DEGREASING TANKS | 


10-110 


10. 
11. 


12. 


AMER 
O}F 


| INDUSTRIAL HYGIED 


Industrial Ventilation 


Solvent vapor degreasing refers to boiling liquid cleaning systems utilizing trichloroethylene, 


perchlorethylene, methylene chloride, freons® or other halogenated hydrocarbons. Cleaning action 
is accomplished by the condensation of the solvent vapors in contact with the work surface 
producing a continuous liquid rinsing action. Cleaning ceases when the temperature of the work 
reaches the temperature of the surrounding solvent vapors. Since halogenated hydrocarbons are 
somewhat similar in their physical, chemical and toxic characteristics, the following safeguards 
should be provided to prevent the creation of a health or life hazard: 


1. 


Vapor degreasing tanks should be equipped with a condenser or vapor level thermostat to keep 
the vapor level below the top edge of the tank by a distance equal to one-half the tank width or 
36 inches, whichever is shorter. 


Where water type condensers are used, inlet water temperature should not exceed 80 F and 
the outlet temperature should not exceed 110 F. For some solvents, lower water temperatures 
may be required. 


Degreasers should be equipped with a boiling liquid thermostat to regulate the rate of vapor 
generation, and with a safety control at an appropriate height above the vapor line to prevent 
the escape of solvent in case of a malfunction. 


Tanks or machines of more than 4 square feet of vapor area should be equipped with suitable 
gasketed cleanout or sludge doors, located near the bottom, to facilitate cleaning. 


Work should be placed in and removed slowly from the degreaser, at a rate no greater than 11 
fom, to prevent sudden disturbances of the vapor level. 


CARE MUST BE TAKEN TO PREVENT DIRECT SOLVENT CARRYOUT DUE TO THE 
SHAPE OF THE PART. 


Maximum rated workloads as determined by the rate of heat transfer (surface area and specific 
heat) should not be exceeded. 


Special precautions should be taken where natural gas or other open flames are used to heat 
the solvent to prevent vapors* from entering the combustion air supply. 


Heating elements should be designed and maintained so that their surface temperature will not 
cause the solvent or mixture to breakdown* or produce excessive vapors. 


Degreasers should be located in such a manner that vapors” will not reach or be drawn into 
atmospheres used for gas or electric arc welding, high temperature heat treating, combustion 
air or open electric motors. 


Whenever spray or other mechanical means are used to disperse solvent liquids, sufficient 
enclosure or baffling should be provided to prevent direct release of airborne vapor above the 
top of the tank. 


An emergency quick-drenching facility should be located in near proximity to the degreaser for 
use in the event of accidental eye contact with the degreasing liquid. 


*Electric arcs, open flames and hot surfaces will thermally decompose halogenated hydrocarbons to toxic and corrosive 
substances (such as hydrochloric and/or hydrofluoric acid). Under some circumstances, phosgene may be formed. 


ICAN CONFERENCE 
GOVERNMENTAL 


SOLVENT VAPOR DEGREASING 


ENISTS 


\. 


[nouns Vow OO OT TE 


10.75 PAINTING OPERATIONS 


Application of industrial paints and coatings usually is 
accomplished by one of three techniques: air-atomization, 
electrostatics, or airless methods. Potential health hazards 
exist from exposure to solid and liquid aerosols as well as to 
solvent vapors. In addition to the airborne exposures, hazards 
include the use of flammable and combustible liquids and the 
accumulation of flammable paint residues. Fire safety and 
proper electrical wiring are important concerns in most paint 
applications. 75.1, 10.75.2) 


Control of airborne pollutants by ventilation may be ac- 
complished through the use of spray booths such as shown in 
VS-75-01 and VS-75-04. The typical booth is a partial enclo- 
sure of sheet metal construction with openings for conveying 
the work piece into and out of the booth. Several factors are 
important in the performance of these booths. Booth depth is 
critical; spray rebound may escape from shallow booths and 
increase exposures. The size of the booth is governed princi- 
pally by the size of the object being coated. Sufficient space 
must be provided to permit air flow on all sides of the object, 
to provide room to work, and to enable the air to enter the 
booth ina smooth, controlled manner without excessive wrap- 
around. In some cases, downdraft booths may be employed 
when large objects are painted. 


In air-atomization applications, the most common spray 
technique, it is important to use the minimum air pressure 
needed to accomplish the task. Excess air pressure results in 
increased dispersion of the paint and overspray as well as poor 
work quality. 


Airless application results in aerosols with fewer particles 
in the respirable range. One study“) suggests that approxi- 
mately 20% of the particles in air-atomization applications are 
less than 12 microns while airless methods produce aerosols 
with only 2% less than this value. The larger aerosols pro- 
duced by the airless technique will deposit more efficiently 
on the work piece, due to impaction, than the smaller particles 
produced by the compressed air method. 


Electrostatic applications result in more efficient deposi- 
tion of paint aerosols due to electrostatic forces. As a result, 
ventilation air flow requirements for control of electrostatic 
applications tend to be lower than for compressed air methods. 


Many spray booths are equipped with disposable particu- 


Specific Operations 10-111 


late filters which become loaded over time and result in 
increased pressure loss. This loss eventually can reduce air 
flow to unacceptable levels and, hence, system performance 
must be monitored. Water wash systems are available for 
cleaning particulate matter from the exhausted air but do little 
for solvent vapors. Fan selection is an important component 
of a spray booth installation. Often the fan is an integral part 
of the system when purchased and may be installed in a 
different configuration than originally designed. This can 
result in reduced air flow, particularly if additional system 
resistance is encountered in the actual installation.(°7*. 


Work practices remain an important aspect of controlling 
exposure to paint aerosols and solvent vapors. The worker 
should not stand downstream of the object being sprayed. A 
turntable can help to facilitate easy access to all sides of the 
object without the worker having to move. Extension arms on 
spray guns should be employed for hard-to-reach cavities. 
Proper location of the booth with respect to replacement air 
and obstructions is essential. Locating booths in corners or 
near disruptive air currents can defeat the protection of these 
hoods. Poor location of the booth may result in turbulent air 
flow which may reduce the protection provided by the booth. 


Respiratory protection may be required in applications 
using toxic materials. This includes heavy metal pigments and 
organics such as isocyanates in urethane paints and amines in 
epoxy paints. 


REFERENCES 


10.75.1 National Fire Protection Association: Flammable 
and Combustible Liquids Code. No. 30. NFPA. Bos- 
ton, MA (1990). 


10.75.2 National Fire Protection Association: National Elec- 
tric Code. No. 70. NFPA. Boston, MA, (1990). 


10.75.3 National Institute for Occupational Safety and 
Health: An Evaluation of Engineering Control Tech- 
nology for Spray Painting. DHHS (NIOSH) Pub. 
No. 81-121; NTIS Pub. No. PB-82-162-264. Na- 
tional Technical Information Service, Springfield, 
VA (1981). 


10.75.4 Burgess, W.A.; Ellenbecker, M.J.; Treitman, R.D.: 
Ventilation for Control of the Work Environment. 
John Wiley and Sons, NY (1989). 


10-112 Industrial Ventilation 


1. Split baffle or filter: 2. Angular baffle 
B = 0.75 D E=D+6 
Baffle area = 0.75 WH Baffle area = 0.40 WH 
For fitter area, see note 2 For filter area, see note 2 


Air spray paint design data. 
Any combination of duct connections and baffles may be used. Large, deep booths 
do not require baffles. Consult manufacturers for water-curtain designs. Use explosion 
proof fixtures and a non-sparking fan. Electrostatic spray booth requires automatic 
high~voltage disconnects for conveyor failure, fan failure or grounding. 


Walk-in booth Operator outside booth 
W = work size + 6’ work size + 2’ 
H = work size + 3’ (minimum = 7’) work size + 2’ 

C = work size +, 6” 0.75 x larger front. dimension 
Q = 100 ctm/tt? booth cross section 100 ~ 150 cfm/ft* of open 
May be 75 cfm/ft*© for very area, including conveyor 
large, deep booth. Operator may openings. 

require a NIOSH cerified respirator. 

he= 1.78 VPs + 0.50 VFy (baffles) 

he = Dirty filter resistance + 0.50 VPq (filters) 
Duct velocity = 2000 fpm 


Airless spray paint design Notes : 1. Baffle arrangements shown are for 
Q = 60 ctm/tt2 booth cross air distribution only. 
section, walk-in booth . Paint arresting filters usually selected 
60 — 100 cfm /fi2 of for 100 — 500 fpm, consult manu-— 
total open area, operator facturer for specific details. 
euieide ef ‘boath . For construction an safety, consult 
NFPA — See Reference 10.75.1. 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 


INDUSTRIAL HYGIENISTS EF Te aaa 


LARGE PAINT BOOTH 


Specific Operations 10-113 


1. Solid baffle . Angular baffle 3. Split baffle or filters 
B.-= 0275: D B=D + 6 B=D+ 6 
Baffle area = 0.60 WH Baffle area = 0.60 WH Baffle area = 0.75 WH 


Air spray design data 
Any combination of branch ducts and baffles may be used. 
= work size + 12° 
= work size + 12° 
= 0.75 W or H, whichever is larger 


W 
H 
C 
Q 


200 cfm/ft? (200 WH) - for face area up to 4 ft? 
150 cfm/ft* - for face area over 4 ft? 
he = 1.78 VP, + 0.25 VPy (baffles) 
= dirty filter resistance + 0.25 VPy (filters) 
Duct velocity = 2000 fpm 


Airless spray paint design data 
Q = 125 cfm/ft? (125 WH) —- for face area up to 4 ft? 
= 100 cfm/ft? - for face area over 4 ft? 


Notes: 1. Baffle arrangements shown are for air distribution only. 


2. Paint arresting filters usually selected for 100 — 500 fpm, consult 
manufacturer for specific details. 


For construction and safety, consult NFPA (Reference 10.75.1). 

AMERICAN CONFERENCE 
OF GOVERNMENTAL 

| INDUSTRIAL HYGIENISTS 5c} aN (IG See 


SMALL PAINT BOOTH 


10-114 Industrial Ventilation 


-Slip sleeve in duct. 


Use window or 


opening at opposite i, 


end of inlet. 


Q = 50 cfm/ft*® of cross~sectional trailer area 
he = 0.25 VFy 
Minimum duct velocity = 2000 fpm 


Paint arresting filters usually selected for 100-500 fpm, 
consult manufacturer for specific deiails. 


For construction and safety, consult NFPA, Reference 10.75.1. 
. Operator must wear an appropriate, NIOSH certified respirator. 


AMERICAN CONFERENCE | TRAILER INTERIOR 


IP DATAIT TAL 
|} OF GOVERNMENTAL = [ Diten® Dfanaie 
J INDUSTRIAL HYGIENISTS [ame 3 PORE VSP 5—09 


Specific Operations 10-115 


Paint arresting filters — 
| in door 


ilters in door desirable 


PLAN VIEW 


\ 
ELEVATION Latch to close doors ~~ 
tightly 
Q = 100 cfm /ft? of cross sectional area ; 
(When W x H is greater than 150 ft*, Q = 50 cfm/ft* ) 


Ne = 0.50 VPq plus resistance of each filter bank when dirty 
Minimum duct velocity = 2000 fpm 


Notes: 
1. Exhaust fan interlock with make-up air supply and compressed 
dir to spray gun is desirable. 


2. Paint arresting filters usually selected for 100-500 fpm. 
Consult manufacturer for specific details. 


3. For construction and safety, consult NFPA, Reference 10.75.1. 
4. For airless spray painting use 
Q = 60 cfm /ft? of cross section area. 


AMERICAN CONFERENCE ily DRIVE THROUG. 
: SPRAY PAINT BOOTH 


: OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


PriGURE 7 se a) 77 


10-116 industrial Ventilation 


sa \ 


\ 


Air filters in doors desirable To fan and discharge (fan should —— 
have inspection door) 


PLAN VIEW 
100 cfm/ft? of cross sectional area 
(when W x H is greater than 150 ft*, Q = 50 cfm/ft?) 
he = 0.50 VPy plus resistance of each filter bank when dirty 
Minimum duct velocity = 2000 fpm 
Paint arresting filters to be sized for 100-500 cfm/ft* 
of filter. Consulf manufacturer for specific details. 


<> 


Le Alternate exhaust duct 


Paint arresting filters 


= duct diameter + 6° 


ELEVATION 
Typical filter installation 


Note: For airless spray painting use 
Q = 60 cfim/ft* of cross-sectional area 


For construction and safety 
consull NFPA code (Reference 10.75.1). 


AMERICAN CONFERENCE | PAINT BOOTH 


OF GOVERNMENTAL es 


INDUSTRIAL HYGIENISTS a I URE Vo 75. 


Specific Operations 10-117 


Locate takeoffs 15 feet on center 

Q = 50 cfm/ft* drain board area, 
but not less than 100 fpm indroft 
through openings 

h, = 0.25 VP, 


e 


Minimum duct velocify = 2000 fpm 


For best 


results enclose 


drainboard as a tunnel 


ee 


- 1000 fpm maximum 
plenum velocity 


Q = 125 cfm/ft* of tank and drainboard area 
Slot velocity = 2000 fpm 

he= 1.78 VP; + 0.25 VP, 

Minimum duct velocity = 2000 fpm 


For air drying in a room or enclosure, see Chapter 2 for 
dilution ventilation required. 


For construction and safety, consult NFPA codes, Reference 10.75.1 


AMERICAN CONFERENCE 
| OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


DIP TANK 


TFIGURE. 


10-118 industrial Ventilation 


Extend as low as possible 
to clear work 


Side baffles 
are desirable 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


LOT TYPE 


Le 

=) 

Q = 100 cfm/ft* door pilus 1/2 
Products of combustion 


Minimum duct velocity = 2000 fpm 
he = 1.0 VP, + 0.25 VPy 


Size plenum for 
1000 fpm maximum 


Slot on three sides with 


Vs = 2000 fpm. 


Locate on inside or outside of door. 


CANOPY TYPE 


Q = 200 ctm/ft4 of hood face 
plus 1/2 products of combustion 


he = 0.25 VPq 
Duct velocity = 2000 fpm 


Notes: 


For dryers, include rate of 

water vapor liberated. 

For flammable solvent drying 

refer to Chapter 2, General 
Industrial Ventilation . 

Hoods at each end of oven. Reduce 
size of doors as much as possible. 
Separate vent must be added for 
products of ccmbustion. 


For construction and safety, consult 
NFPA code (Reference 10.75.1). 


DRYING OVEN VENTILATION 


~~ «YRIGURE VS 75 20 y 


Q(exh) = 10 to 12 air changes/hour 
Q(supply) = 10 to 12 air changes/hour 
Balance room slightly negative 

Q(exh) = 1.05 Q(supply) 

Entry loss = 1.78 VPs5+ 0.05 VPp 
Minimum duct velocity = 2000 fpm 


Stack velocity = 3500 fpm 


Floor line ——- Ee 


fF For construction and safety consult NFPA Code 
pe 
AMERICAN CONFERENCE 

: OF GOVERNMENTAL 

7 INDUSTRIAL HYGIENISTS 


Specific Operations 


Paint mix 
pot (typ.) 


PAINT MIX 
STORAGE ROOM 


10-119 


10-120 Industrial Ventilation 


10.80 MECHANICAL SURFACE CLEANING AND 
FINISHING 


Mechanical surface cleaning is generally used to clean a 
surface in preparation for painting, welding, or other opera- 
tions. The surfaces may be coated with paint, rust, or oxida- 
tion; plated with other metals; or covered with molding sand, 
inorganic, organic, or biological matter. Mechanical cleaning 
may be accomplished by abrasive blasting, wire wheels, sand 
paper/sanding belts, grinding wheels, or use of abrasive chips 
in tumbling mills. The capture velocity needed to entrain large 
particles is often very high and the collection hood must be 
positioned so the materials are directed toward the hood. A 
minimum duct transport velocity of 3500 fpm is needed but 
4000 to 5000 fpm is recommended. A hood that encloses as 
much of the operation as practical is desired. The toxicity of 
the material removed must be considered when cleaning 
mechanically. Complete enclosures may be used or the 
worker may need to wear a respirator in addition to using local 
exhaust ventilation. 


For many grinding, buffing and polishing operations, regu- 
lations from the Occupational Safety and Health Administra- 
tion (OSHA) °*") and National Fire Protection Association 
(NFPA)(°8°2 may apply. 


10.80.1 Abrasive Cleaning: VS-80-01, -02, and -03 show 
suggested designs for abrasive blasting and tumbling mills. A 
supplied air respirator must be used in abrasive blasting 
rooms. 


10.80.2 Grinding: Mechanical surface finishing uses or- 
ganic bonded wheels, cones, saws or other shapes rotating at 
a high rate of speed to smooth a surface; reduce an object or 
part in size; or perform other operations. As the object is being 
surfaced or finished, metallic particles are removed and leave 
the object at a high speed. In addition, the abrasive wheel is 
reduced in size and generates particles that must be controlled. 
Frequently, grinding is accomplished using fluids to keep the 
parts cool. This cooling fluid will be emitted as an aerosol or 
mist and needs to be controlled and provisions must be made 
in the duct to drain off the liquids that accumulate. 


The hood used to capture the particles should enclose the 
operation as much as possible and be positioned to take 
advantage of the velocity and direction of the particles as they 
are generated. Design specifications for grinding and surfac- 
ing operations are shown in VS-80-10 through VS-80-19. 


10.80.3 Buffing and Polishing: The same principles ap- 
ply for buffing and polishing as for grinding and surfacing. 
The buffing wheel or belt should be enclosed as much as 
practical and positioned to take advantage of the centrifugal 
force of the particles as they leave the wheel or belt. The 
minimum duct velocity for the generated particles is 3500 fpm 
and 4500 fpm if the material is wet or sticky. Since many 
varieties of metals and alloys are buffed and polished, it is 
extremely important not to mix ferrous and non-ferrous met- 
als in the same exhaust systems (see NFPA codes),.(1°8°2) 
VS-80-30 through VS-80-35 show suggested designs for 
buffing and polishing. 


REFERENCES 


10.80.1 U.S. Department of Labor, Occupational Safety and 
Health Administration: 29 CFR.1910. US- 
DOL/OSHA, Washington, DC (1970). 


10.80.2 National Fire Protection Association: National Fire 
Codes—in particular NFPA-65 (Processing and Fin- 
ishing of Aluminum); NFPA-68 (Guide for Explo- 
sion Prevention Systems); NFPA-77 (Practice on 
Static Electricity); NFPA-91 (Installation of Blow- 
ers and Exhaust Systems for Dust, Stack and Vapor 
Removal or Conveying); NFPA-480 (Storage, Han- 
dling and Processing of Magnesium); NFPA-481 
(Production, Processing, Handling and Storage of 
Titanium); NFPA-482 (Production, Processing, 
Handling and Storage of Zirconium); and NFPA- 
561 (Manufacture of Aluminum and Magnesium 
Powder). NFPA, Quincy, MA. 


10.80.3 National Institute for Occupational Safety and 
Health: Recommended Industrial Ventilation 
Guidelines. DHEW (NIOSH) Pub. No. 76-162; 
NTIS Pub. No. PB-266-227. National Technical In- 
formation Service, Springfield, VA (1975). 


10.80.4 American Foundrymen's Society, Inc.: Foundry 
Ventilation Manual. AFS, Des Plaines, IL (1985). 


10.80.5 National Institute for Occupational Safety and 
Health: Ventilation Requirements for Grinding, 
Buffing and Polishing Operations. DHEW (NIOSH) 
Pub. No. 75-107; NTIS Pub. No. PB-277-332. Na- 
tional Technical Information Service, Springfield, 
VA (1975). 


Specific Operations 10-121 


> 
3500 fpm, minimum ——— 2 as To dust collector 


— Air inlets, baffled " 


aS - Ea oe eRe es eee 


SECTION THROUGH TYPICAL ROOM 


60-100 cim/ft2 of floor for downdraft with typical choice 80 cfm /ftA. 
100 cfm/ft4 of wall for crossdraft- 


Q 
Q 


i 


lt 


Lower control velocities may be used depending on toxicity of the contaminant, 
object and blasting media and the size of the blasting room. 


Notes: 1. The above ventilation is for operator visibility and to control escape 
of contaminants into adjacent work areas. 


2. Operator in an abrasive blasting room is required to wear appropriate 
NIOSH certified respiratory protection. 


3. For rotary tables use 200 cfm/ft* of total opening (taken without 
curtains). 


For blasting cabinets see VS—80~-02. 


B Reference 10.80.3 and 10.80.4 


ABRASIVE BLASTING 
ROOM 


~PFIGURE 


AMERICAN CONFERENCE 
: OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


10-122 Industrial Ventilation 


Exhaust System 


SH 


Screened viewing - 2 ood, ell ~ Door with 
window 9 -———~-— | dust tight 
| gasket 


Rubber gloves 
attached to 
cabinet. ———~--— 


~ Return to blasting hopper fer reuse 


20 dir changes per minute 

At least 500 fpm inward velocity at all openings 
Minimum duct velocity = 4000 fpm 

If cabinet has self-contained dust collector, 
consult manufacturer for losses. 

Ne = 1.0 VP, 


| AMERICAN CONFERENCE 
OF GOVERNMENTAL 


: | ABRASIVE BLASTING CABINET 
INDUSTRIAL HYGIENISTS | 


02-91 [rout VS 50-02 | 


a End bell 


SECTION THROUGH HOLLOW 
TRUNNION TUMBLER 
Minirnum duct velocity = S000 fpm 


Entry loss (hg) depends on design and 
typically ranges from 3 to 9 ‘wg. 


Specific Operations 10-123 


Hinged access ae 


Air slot velocity = 
400 for. minimum 


STAVE MILL 
(END SECTION) 


Minimum duct velocity = 3500 fpm 
he = 0.25 VR 


Square mill 


| 
| 


j 


| Trunnion | 


430 


24 


31 


2000 > 2] 


* Low-—loss designs have large air inlet openings i 


Holes in end discs are sized for velocities of 1250 — 1800 fpm. 


** For lengths over 72’, increase exhaust rate proportionately 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


TUMBLING: MILLS 


10-124 Industrial Ventilation 


r~ Adjustable tongue (keep adjusted to not 
| more than 1/4 from wheel ) 


1” clearance } 
Se oe q? TT 
6 es ; 


i 


SEEN a er cee al 


SPECIAL TOOL REST 
Reference 10.10 


Ny 


Va Chip trap if desired 
EXHAUST FLOW RATES, cfm 


Wheel diam. Wheel Good Poor 
width 
_enclosure* | enclosure | 


*Special hood and tool rest as shown, no more than 25% of the wheel exposed. 
Minimum duct velocity = 4000 fpm 


ne = 0.65 VFy for straight take-off 
= 0.40 VFy for tapered take-off 


GRINDING WHEEL HOOD — 
SURFACE SPEEDS 
ABOVE 6500 sfpm 


AMERICAN CONFERENCE 
| OF GOVERNMENTAL 
INDUSTRIAL 


HYGIENISTS ae peo 


Specific Operations 10-125 


— Adjustable tongue (keep adjusted to not 
more than 1/4" from wheel) 


Bue Chip trap 
ae on if desired 


EXHAUST FLOW RATES, cfm 


Wheel diam Wheel Good fp Poor ; 
inches | wiih enclosure* enclosure 
Up to 5 1 220 220 
“Pe TOs. es 220 | 300 | 
10 to 14 2 300 | 500 
PAO Ae 2 390 610 
P 46t00 | 3) 500 | 940 
20 to 24 4 610 880 
24 to 30 i 5 | 880 | 1200 
30 to 36 6 i200-- |. - 4600. 4 


* No more than 25% of wheel exposed. 
Minimum duct velocity = 4000 fpm 


he = 0.65 VFy for straight takeoff 
0.40 VPy for tapered takeoff 


GRINDING WHEEL HOOD — 
SURFACE SPEEDS 
BELOW 6500 sfpin 

pave UP GI] | 02 oy. a | FIGURE Vio 80 ee 71 | 


AMERICAN CONFERENCE 
: OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


10-126 Industrial Ventilation 


Grinding wheel ~~ 


Metal strip 


tt 


0.043 V, (10x" + A) 

mum duct velocity = 3500 fpm 
0.25 VPy 

distance from hood face to center of wheel, ft 
nood face area, ft 

Wheel Speed, surface feet per min. (SFM) 

m7 (D/12) R 

diameter in inches 
rpm of grinding wheel 


Wo We th Te th 


Reference 10.80.5 : 


AMERICAN CONFERENCE 
| OF GOVERNMENTAL = | | 
J INDUSTRIAL HYGIENISTS Em 7 Tam VOB OTS I 


SURFACE GRINDER 


Specific Operations 10-127 


~ Slip and swivel joint 


7 a Supports 


-—- Cut to suit around frame 


_-—~ Use canvas or rubber baffles 
fo surround disc as far as 


possible 


Disc _ 


Diameter 


| Disc diameter {Duct diameter | cfm_ 
Up to 20° 
20° to 30° 
30° to 53° 


53” to 72" 


t 


i 


= 4000 fpm 


Minimum duct velocity 
= 2000 fpm 


Minimum slot velocity 


a Bes A VE Oa VE 


@ n,+ elbow losses + joint losses 


Seep ee eee eee CORE GRINDER 


OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


10-128 Industrial Ventilation 


Ring attached to hood at 
convenient locations 


Adjustable to 
1 Oo fh fe 0 fii clear grinder 


—| 


—+-— 


be | 


es 


“ Angle of slots to be in relation to rotation 


I 
Noss BN 


EXHAUST FLOW RATE, cfm 


Disc diameter 1/2 or more of disc covered Disc noi covered 
inches ON oN 


oO. 
2 
2 
4 
2 


| Exhaust flow rate, cfm _ Exhaust flow rate, cfm | 


of exhaust outlets around periphery of hood or equal distribution 
provided by other means. 


Minimum slot velocity = 2000 fpm 
Minimum duct velocity = 4000 fpm 
he= 1.0 VP, + 0.5 VFy 


AMERICAN CONFERENCE VERTICAL SPINDLE DISC 
OF GOVERNMENTAL : GRINDER 


| INDUSTRIAL HYGIENISTS paps 5 tae 


Specific Operations 10-129 


— belt conveyor or | 
any other method. 


Disc diameter {Exhaust flow 
inches cfm 
rn Up tO: 19. i. 610 sj) Section A-A 
19 to.25 880 
Ce he a4 ae a Note: If the disc is tightly enclosed by machine 
$0 fo 950 2000 housing, then exhaust from the housing is 
93 to 72 6300 acceptable. 


Minimum duct velocity = 4000 fpm 
he = 0.65 VP, straight take-off 
= 0.45 VPy tapered take-off 


e 


AMERICAN CONFERENCE 
| OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


HORIZONTAL DOUBLE—SPINDLE 
DISC GRINDER 
DATE 02 — 9] | rod FIGURE 80-1 ; 


10-130 Industrial Ventilation 


<- Branch takeoff at top or back. Central location 

~~ multiple branches if several booths are used. 
ee Additional adjoining 
ia booths if needed. 


4S’ slope —-~\ 


Booth encloses grinder ‘ |-- Keep width as small 


frame and suspension. -——~ as practical. 


Grinder to operate in 
close fo face opening. - 


For a large opening, 4’ to 6’ wide 
Q = 150 cfm/ft* of opening 


For a small opening, 2’ to 2’- 
grinder in front 
Q = 200 cfm/ft* of opening 


Minimum duct velocity = 3500 fpm 
he = 0.25 VFy 


: Small local exhaust hoods mounted behind 
grinder wheel may trap the stream of sparks, 
but are usually not effective in control of 
airborne dust. 


AMERICAN CONFERENCE Pennie rae 
a ‘ GRINDER 
OF GOVERNMENTAL ee a 


INDUSTRIAL HYGIENISTS Tap S5_o77 ncurses Vo S0-76 | 


Specific Operations 10-131 


Reduce open area 
with baffles 


Booth width to suit ~ 
regular work 


Hinged side doors may be 
opened for longer pieces —~_[*\ 


Saw operates at face 
Sif OO ie ee ie 


~ Rear takeoff 
(best) 


iz~ Hinged cleanout 
Ke door 


Close in area under 
Hej billie jet 


Q = 250 cfm/ft? of open face area 
Minimum duct velocity = 4000 fpm 
he = 0.50 VPy (no taper) 

0.25 VPy (with 45° taper) 


AMERICAN CONFERENCE | ABRASIVE CUT-OFF 
OF GOVERNMENTAL : SAW 


ee 


10-132 Industrial Ventilation 


—- Back and side shields highly 
desirable, enclose sides and 
top to make booth if practical. 


OT 7 Bench top 7 


| { ; \y | 
oe VV 7 
os tl Cleanout doors 45 malts 


or drawers. Ms 


“= Tapered takeoff necessary 


for distribution. 
END VIEW 


Q = 150-250 cfm/ft* of bench area. 
Minimum duct velocity = 3500 fpm 
he = 0.25 VPy 

lf slots are used for distribution 

he = 1.78 VP; + 0.25 VPy 


Notes: lf grinding in a booth, use 100 fom face velocity. 


ile 
2. For downdraft grilles in floor: Q=100 cfm/ft” of working area. 
3. Provide equal distribution. 

4. Provide for cleanout. 


AMERICAN CONFERENCE 
: OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS | 


HAND GRINDING BENCH 


VWS— 50-168 | 


Specific Operations 10-133 


oe eo ae eee eee 


| -- Opening to be 
sized to handle 


3/4 of total air \ 2/3 of duct 
at 1000 fpm 


y 
SSS PPR EP TEP ES Sy aT FUERTE, aT 
a fe eed 


amy 


4 
| 
| 
| 
| 
} 
i 
' 
t 
i 
t 
| 
I 
| 
1 


1/4 of total air at 200 fpm 


Q = 150 cfm/ft* opening 
Minimum duct velocity = 3500 fpm 
he = 0.25 VPy 


AMERICAN CONFERENCE | PORTABLE CHIPPING AND 
|} OF GOVERNMENTAL = | GRINDING TABLE 
| INDUSTRIAL HYGIENISTS [ing——QDTO]—— neue OT OT To I 


10-134 Industrial Ventilation 


as close to wheel as 


Adjustable fongue — keep 1 
panels: poo IK 0.25 D at least. 


Direction of rotation. 


for hard wheel. 
for soft wheel. 


- Trap with cleanout 


possible when desirable. 


Minimum duct velocity : 3500 fpm, 
4500 fpm if material is wet or sticky 


Re 0.65 VPy for straight take-off. 
Ne 0.40 VPqy for tapered take-off. 


Wheel diam. | Wheel width Exhaust flow rate | Exhaust flow rate 
CS ce OES: Ns Seer — 


Good enclosure + “Poor. enclosure 


‘Up to 9 


over 9 to 16 | 


* not more than 25% of the wheel is exposed 
Note : Consult applicable NFPA codes Reference 10.80.2 
Caution : Do not mix ferrous and non-ferrous metals 
in same exhaust system. 


AME RIC AN CONFERE NCE MANUAL 
OF GOVERNMENTAL BUFFING AND POLISHING 


INDUSTRIAL ne ee [" bate Q>-91 peor TReuRE VS— GO-GO | 


Specific Operations 10-135 


~ When main duct is overhead, 
locate preferred take-off 
: location as shown. 
Adjustable tongue ~— keep — 


minimum clearance. 


|-~ Keep side clearance 
] to a minimum. 


Adjustable hopper — 


: 
poy = 
I 


~ Aternate § 
take-off.§f 


Cleanout door ee ge 


ize 


Wheel diam. Wheel width Exhaust flow rate 
inches inches 


For wider wheels than listed, increase cfm with width 
Minimum duct velocity = 3500 fpm 


4500 fpm if material is wet or sticky 
he = 0.40 VPq 


. Consuli applicable NFPA codes. See Reference 10.80.2 
. For titanium, aluminum, and magnesium, eliminate hopper 
use 5000 fom through hood cross-section. 


. Caution : Do not mix ferrous and non~ferrous metals in same exhaust system. | 


AMERICAN CONFERENCE ee . oo geet 
hs : perenne sa bee vi af 7 LATHE 
OF GOVERNMENTAL PG, aang 
INDUSTRIAL HYGIENISTS 


TG ==CF (SLR STORES? 


10-136 Industrial Ventilation 


- Side opening should be minimal 
1/4 maximum desired 


— Adjustable tongue; not more than 1/4” 
from belt 


_.Hinged side panel 
maintenance 


~ Pad 
Cleanout door-~ i 


_ -s ii a in ms at ke ‘ ] < 

i BS a A i y K aa — 

, For heavy dust accumulations 5 | ee 
housing may extend to floor aa 


Belt width pxngust flewewre Exhaust flow rate 
inches |. cfm cfm 


‘Good enclosure *;} Enclosure 


1570 


* Hood as shown; no more than 25% of wheel 
exposed. 
he = 0.40 VPy 
Minimum duct velocity = 3500 fpm, 4500 fpm if wet or sticky. 
Notes: 


1. Consult applicable NFPA codes, 10.80.2 

2. For titanium, aluminum and magnesium eliminate hopper and use 
5000 fpm through hood cross section. 

3. Caution: do not mix ferrous and non-ferrous 
metals in same exhaust system. 


BACKSTAND IDLER 
POLISHING MACHINE’ 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
DUSTRIAL HYGIENISTS 


IN 


Specific Operations 10-137 


Use one branch duct 
/ for each wheel 


__ Hinged access doors for maintenance, 
/ normally closed 


— Slow speed belt conveyor 


Q = 500 cfm/wheel, minimum 
Not less than 250 cfm/ft total open area 
Minimum duct velocity = 3500 fpm, 4500 fpm if 
material is wet or sticky 
he = 1.78 VP; + 0.25 VP, 


Note: 
1. Consult applicable NFPA standards. Reference 10.80.2 
2. Caution: Do not mix ferrous and non-ferrous 

metals in same exhaust system. 
3. Wheel adjustments on outside of enclosure. 
4. For highly toxic material, enclose the return strand 
of the belt conveyor. 


STRAIGHT LINE AUTOMATIC 
) BUFFING 
pare 02-97 Pat V5—80— 33 | 


N CONFERENCE 
OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


AMERICA 


10-138 Industrial Ventilation 


Access door 


Q = 500 cfm/wheel, minimum, 
Not less than 250 cfm/ff total open area 
Minimum duct velocity = 3500 fpm, 4500 fpm if 
material is wet or sticky 
he = 1.78 VP5 + 0.25 VP, 
On small, 2 or 3 spindle machines, one take-off may be used. 
Multiple take—offs desirable. 


Note: 
1. Consult applicable NFPA standards, Reference 10.80.2 
2. Caution: Do not mix ferrous and non-ferrous 

metals in same exhaust system. 


AMERICAN CONFERENCE 
1 OF GOVERNMENTAL 
7 INDUSTRIAL HYGIENISTS | ep 


CIRCULAR AUTOMATIC BUFFING 


VS—80— 34 J 


Specific Operations 10-139 


Sliding tongue 


r. Belt tension 


Side hinged 


| Belt width, inches | 


imum duct velocity 
terial is wet or sticky 
= 0.65 VPqfor straight take-off 
0.45 VPq for tapered take-off 

10.80.2 


Notes: 
Consult applicable NFPA codes Reference 
Do not mix ferrous and non-ferrous 


ibe 
2. Caution : 
metals in same exhaust system 


Min 
ma 
h 

e 


BELT 


METAL POLISHING 


ICAN CONFERENCE 


| AMER 
| OF GOVERNMENTAL — | 
HYGIENISTS DATE O02 a OL ~ EPICURE, 


INDUSTRIAL 


10-140 industrial Ventilation 


10.85 VEHICLE VENTILATION 


The objective of providing ventilation for vehicles in an 
environment is to keep a worker's exposure to toxic exhaust 
fumes and gases below the TLV, both the TWA and STEL, 
or other appropriate standards. This can be achieved either by 
dilution or local exhaust ventilation. 


It is difficult to establish dilution ventilation requirements 
accurately for the operating vehicles in a plant. For an existing 
facility, the designer has the opportunity to measure the 
emission in the field. Standard techniques can be used to 
measure gas flow rates, composition and temperatures, and 
contaminant levels. Using the equations in Chapter 2 and the 
measurements, the dilution rates can be calculated. However, 
it Is not always possible to accurately determine the contami- 
nant generation rate because generation is not uniform. More- 
over, no such data are available to the designer for new 
vehicles. 


The use of dilution ventilation is usually considered only 
after rejection of the source capture concept. Common reasons 
for rejecting source capture (local exhaust) are operating 
interference problems or layout constraints. For lift trucks or 
cars in motion or idling outside of stalls, local exhaust is not 
feasible. Hence the only method for control of health hazards 
is dilution ventilation. 


Over the years, some empirical rates have been developed 
which have been applied successfully to achieve contaminant 
control. The recommended dilution rates based on average 
operating conditions are: 


5,000 cfm/propane fueled lift truck 
8,000 cfm/gasoline fueled lift truck 
5,000 cfm/operating automobile 
10,000 cfm (or more)/operating truck 
100 cfm/horsepower for diesel fueled vehicle 


The above dilution rates for lift trucks apply under the 
following conditions:('85.) 


1. A regular maintenance program incorporating final 
engine tuning through carbon monoxide analysis of 
exhaust gas must be provided. CO concentration of 
gases should be limited to 1% for propane fueled 
trucks; 2% for gasoline fueled trucks. 


2. The periods of lift truck engine operation do not ex- 
ceed 50% of the working day (total engine operation 
of lift truck equal to or less than 4 hours in an 8-hour 
shift). 

3. A reasonably good distribution of air flow must be 
provided. 


4. The volume of space must amount to 150,000 f/lift 
truck or more. 


5. The lift truck is powered by an engine of less than 60 
HP. 


Where actual operating conditions vary from the above, the 
ventilation rate should be increased. On the other hand, me- 
chanical ventilation may not be required in large buildings 
where lift truck operation is intermittent and where natural 
infiltration based on a maximum of one air change/hour for 
the net building volume exceeds the recommended dilution 
ventilation rate. 


The alternative to dilution ventilation is to capture the 
contaminant at the source by installing local exhaust ventila- 
tion. For stationary vehicles in service garages, effective 
systems are shown in VS-85-01 (overhead) and VS-85-02 
(under floor). The systems should be connected directly to the 
vehicle exhaust and should terminate outdoors above the roof. 
The design procedure outlined in Chapter 5 must be followed. 
For friction loss data of flexible ducts, manufacturers should 
be contacted. As with all flexible systems, the length of 
flexible duct must be minimized, and non-collapsible duct 
should be used. Unnecessary and/or sharp bends should be 
avoided. Exhaust requirements for automobiles are shown in 
VS-85-02 and for diesel engines in VS-85-03. 


The requirements for parking or storage garages should be 
based on short-term exposure of drivers to exhaust emissions 
when entering or departing. A continuous supply of 500 cfm 
fresh air/parking space should be adequate. Additional venti- 
lation may be required if there are long periods of engine 
idling (winter warm-ups, loading, etc.) or if the general traffic 
pattern is such that clusters of vehicles arrive or depart. 


Attendant booths of parking garages should be pressurized 
with a supply of fresh air from uncontaminated sources. 


For indoor loading docks, continuous supply of 2 cfm/ft? 
of dock area should be adequate where truck motors are shut 
off except when entering or leaving the dock. 


REFERENCES 


10.85.1 Hama, G.M.; Butler, Jr., K.E.: Ventilation Require- 
ments for Lift Truck Operation. Heating, Piping and 
Air Conditioning (January, 1970). 


10.85.2 Goldfield, J.; Sheehy, J.W.; Gunter, B.J.; Daniels, 
W.J.: An Affordable Ventilation Control] for Radia- 
tor Repair Shops. Ventilation ‘91: 3rd International 
Symposium on Ventilation for Contaminant Con- 
trol. American Conference of Governmental Indus- 
trial Hygienists, Cincinnati, OH (1993). 


Specific Operations 10-141 


Stackhead — See Fig. 5-31 


cme Main duct: 
\ Plenum design best — size for 2000 fpm maximum 
or design as in chapier 5. 


ca Oc 


ee lexible joi 
—==- Hose can be : 10’-12’ from é-- All joints soldered 
counterweighted floor 


For dual tailpipes 

use one hose with "Y’ 
or use two outlets per 
stall. 


cfm/vehicle | Flexible duct seduce 
| diam. connection _ 


Over 200 


_Diesel trucks” See VS— 85— (05 _ 


On dynamorneter test rolls 
Automobiles and light duty trucks = 350 cfm 
Heavy duty trucks = 1200 cfm minimum 


For friction loss of flexible duct; consult manufacturers’ data. 


See VS-85-02 


AMERICAN CONFERENCE | SERVICE GARAGE VENTILATION 
OF GOVERNMENTAL alee 


INDUSTRIAL HYGIENISTS [RE a | 


10-142 Industrial Ventilation 


Note : In ventilating a garage use eilher the overhead or under floor system. 
Exhaust to be discharged above roof. 


To fan and ~~. 
discharge above ~ 
roof 


xy Double or single floor plates suitable. 


wa 


“ 


Flex duct to -» 


ee - \Self-closing floor plates desirable. : 
dilpipe a a 


J 


Along ceiling of floor below, or in trenc = Size main J 
If in trench, drain tile with cemented 2000 fpm <-Sump or dry 
joints is suitable. Must be sloped and less. well 

drained for flushing. 


UNDER FLOOR SYSTEM 
EXHAUST REQUIREMENTS* 


Type cfm poe Venicis | Flex duct ID ae. 


Aitompbiles. one icke uP 46 200 hp 


AGlom spies and trucks over 200 hp 


| 
{ 
i 
SSE eee vee er eee aoe mits omens eRnee ane aol 
t 
{ 
| 


Diesel 


See VS-85-03 


* On dynamometer test rolls 
Automobiles and light duty trucks = 350 cfm 
Heavy duty trucks = 1200 cfm minimum 


** 3 diam. permissible for short runs with proper fan. 
For friction loss of flexible duct; consult manufacturers’ data. 


Use adapters 

on dual exhausis 
and special 
tailpipes. 


AMERICAN CONFERENCE | SERVICE GARAGE 
| OF GOVERNMENTAL VENTILATION UNDERFLOOR | 
| INDUSTRIAL HYGIENISTS FOS] 


Specific Operations 10-143 


| 
| 
| 


. 


i i 
H H | oedema! eee 


NA — Normally Aspirated | 
TC — Turbo Charged 


i if an aes 
i 
i 
t 


fm 


io 


Exhaust, 


2000 5000 


Engine speed, rpm 


Exhaust, cfm = acfm + 20% excess 
For specific design information request manufacturers 13 mode 


EPA engine bench test 


AMERICAN CONFERENCE | 24HAUST SYSTEM REQUIREMENTS 
| OF GOVERNMENTAL FOR TYPICAL DIESEL ENGINES 
I iate ree nas Tac MiSs UNDER LOAD 


10-144 Industrial Ventilation 


Water tank 


Lift 


: a 


| 
| 
aoa 
Li | 
. Enclosure ss 
anaes opening 

| | 


Elevation View 


Q = 190 cfm/ft? 
opening (L x W) a 
Minimum duct velocity = 3000 fpm a 
he= 0.25 VPq 
L = Height of ventilated opening in mm, 
W = Width of ventilated opening in mm. 


Notes: 1, Enclosure can be made of metal or fire resistant curtain 
material 


: Reference: 10.85.3 


AMERICAN CONFERENCE | VENTILATED BOOTH FOR 
f OF GOVERNMENTAL = | RADIATOR REPAIR SOLDERING | 
INDUSTRIAL HYGIENISTS ETT : 


10.90 WELDING AND CUTTING 


The purpose of welding and cutting ventilation is to control 
gases, fumes, and particulate generated during the welding 
and cutting operations. 


10.90.1 Hazards: The generation rate of fumes and gases 
varies with the composition of the base metal, fluxes, and 
fillers, and with the rate and depth of welding. Exposure to 
the welder varies with the generation rate, duration and fre- 
quency of operations, work practices (particularly distance of 
the plume from the breathing zone), and the effectiveness of 
ventilation. 


Contaminants from welding may include: 


1. Fume from the base metals and filler or electrode 
metals. 


2. Fume from coatings (e.g., zinc oxide from galvanized 
surfaces, thoria from T.I.G. welding, and fluorides and 
NO), from electrode coatings). 


3. Ozone due to ionization of oxygen by the ultraviolet 
light from arc welding. 


4. Carbon monoxide from ultraviolet effects on carbon 
dioxide in shield gas. 


5. Shield gases such as carbon dioxide, helium and argon. 


6. Fluoride gases and other thermal decomposition prod- 
ucts of fluxes and electrode coatings. 


7. Flammable gases such as acetylene. 


There are welding tasks that present enhanced hazards such 
as welding on materials containing or contaminated with 
heavy metals or welding in the presence of flammable vapors 
or halogenated hydrocarbons. If such welding is required, 
extraordinary precautions must be taken on a case-by-case 
basis. Even in the absence of such hazard materials, any 


Specific Operations 10-145 


welding operation in a confined space is potentially lethal and 
requires continuous and copious dilution ventilation. 


10.90.2 General Recommendations: 


1. Choose hood designs in the following descending 
order of effectiveness: enclosing hoods; vacuum noz- 
zles; fixed slot/plenum hood on a worktable or rectan- 
gular hood fixed above a worktable; moveable hood 
above a worktable; moveable hood hanging freely or 
overhead canopy; dilution ventilation. 


2. Integrate planning for ventilation systems with plan- 
ning for materials handling. 


3. Place welding curtains or other barriers to block cross- 
drafts. 


4. Install turntables, work rests, and other aids to improve 
utilization of the hoods. 


5. Avoid recirculating filtered air from welding hoods 
back into occupied spaces unless the welding is low 
hazard and produces low quantities of gaseous con- 
taminants. 


6. Face velocity for enclosing hoods should be 100-130 
fpm with the higher values used for poor conditions 
such as high cross-draft velocities. 


7. Capture velocity for non-enclosing hoods should be 
100-170 fpm with the higher values used for poor 
conditions such as high cross-draft velocities and with 
higher hazard levels. 


Enclosing hoods are by far the most effective in controlling 
welding contaminants; however, they restrict access and force 
reconsideration of material and product handling. Capturing 
hoods are less effective than enclosures but for low hazard 
conditions can be adequate if properly used. 


10-146 


Industrial Ventilation 


4S’ taper angle 


I Slots-size for 2000 fpm 


Baffles are 
desirable 


Maximum plenum velocity 
1/2 slot velocity 


Q = 350 cfm/ft of hood length 


Hood length = required working space 
W f 


= 24 maximum, if W>24 see chapter 3 


Minimum duct velocity = 2000 fom 


where local exhaust can not be used: 


General ventilation, 


Cim/welder A. For open areas, where welding fume can 
rise away from the breathing zone: 
cfm required = 800 x Ib/hour rod used 


For enclosed areas or positions where fume 
does not readily escape breathing zone: 
cfm required = 1600 x Ib/hour rod used 


er airflows are necessary and operator 


For toxic materials hig 
may require respiratory protection equipment. 


Other types of hoods 
Local exhaust: See VS—90-02 
Booth: For design see VS-90-30 


Q = 100 cfm/ft? of face opening 
MIG welding may require precise air flow control 


CAN CONFERENCE | ING VENTILA 
GOVERNMENTAL — | BENCH HOOD : 
AL HYGIENISTS Ke Sy eo 


Overhead 
support 


= 


Swivel 


PLAIN DUCT 


RATE OF EXHAUST 


B cfm cone, cfm 


Up to 6 335 250 
6-9 | 755 | 560 
9-12 | 1335 | 1000 

1 Face velocity = 1500 fpm 


) Minimum duct velocity = 3000 fpm 
f Plain duct entry loss = 0.93 VPq 
f Flange or cone entry loss = 0.25 VPq 


EF Notes: 

1. Locate work as close as 

/ possible to hood, 

f 2. Hoods perform best when located 
L to the side of the work. 

#3. Ventilation rates may be 

E inadequate for toxic materials. 

E 4. Velocities above 100-200 fpm 

H may disturb shield gas. 


AMERICAN CONFERENCE 
: OF GOVERNMENTAL 
|} INDUSTRIAL HYGIENISTS 


To exhaust 


CONE HOOD 


10-147 


Specific Operations 


Cleanout ' 


FLANGED HOOD 


GENERAL VENTILATION, where local exhaust cannot : 


be used : Rod, diam. | cfm/welder 
5/32 1000 
3/16 1500 | 
ee 3500 : 
3/8 4500 


OR 


A. For open areas, where welding fume can rise|j 
away from the breathing zone: 
cfm required = 800 x Ilb/hour rod used 
B. For enclosed areas or positions where fume 
does not readily escape breathing zone: 
cfm required = 1600 x Ib/hour rod used 


For toxic materials higher airflows are necessary 
and operator should use respiratory protection 
equipment. 


Other types of hoods 
Bench, see VS—-30-01 
Booth, for design see VS—390—-30 


WELDING VENTILATION 
MOVABLE EXHAUST HOODS 


10-148 Industrial Ventilation 


BA tNeolaceable 
B filter media 


WCar silhouette> 


Flow line 


= 1000 to 1200 cfm/linear ft of booth 


filter Joss + 0.5 VPq 
Minimum cuct velocity = 3500 fpm 


| AMERICAN CONFERENCE | 
[| OF GOVERNMENTAL | 
| INDUSTRIAL HYGIENISTS | 


——- Window 


XN 


DD Grites 72” 
i 


Grith 2 to 30° high & 

/ Desian for 300-500 fpm 
| __. Exhaust 
: stac S 


| 
Ze 
t a 1 


| 
be 


WELDING BOOTH 


Specific Operations 10-149 


6 foot center — 
to center 


c~ Slots sized for maximum 
— fom 


Cleanout ——!~ 
doors 


——~ Enclose base 
of bench 


Q = 150 cfm/ft? of gross bench area 
Minimum duct velocity = 4000 fpm 
he = 1.78 VPs + 0.25 VPy 


AMERICAN CONFERENCE Bac tae, ee 
a en TORCH CUTTING. VENTILATION 


10-150 Industrial Ventilation 


172 a 


Slot velocity = 2000 fpm 
PA eeig weld Hood located within 4” 
from source to maintain 
200 fpm capture velocity 


: To exhaust blower a : 
: Rigid pipe 


Flexible 
weld hose 


Swivel 


7 


Slot nozzle 
see detail above 


Rotation i 


ra Floor line 


he= 1.78 VR+ 0.25 VPy 
Minimum duct velocity = 3500 fpm 


| AMERICAN CONFERENCE | 
| OF GOVERNMENTAL 


| INDUSTRIAL HYGIENISTS [ae—QS—OT aE STOO OO SOT 


ROBOTIC APPLICATION 


Specific Operations 10-151 


Slot velocity = 2000 fpm— 


Face open—~ | 


—~ grille top work bench 


METALLIZING BOOTH 


Non-toxic: Q = 150 cfm/ft? face area Toxic: Provide appropriate NIOSH 
: certified respirator 
Q = 200 cfm/ft? face area 


Minimum duct velocity = 3500 fpm 
he = 1.78 VPs + 0.25 VPy 
Small lathe, etc., may be mounted in booth 


~ Gun (on tool post) { Flex duct to allow 
\ | movement full length 


{ 
x i of work 
| | 7 


~ SSe 
“clear lathe rail. Hood may be 


LOCAL HOOD connected to move with tool rest. 


Note: Local hood may not be satisfactory for spraying toxic metals. 
Q = 200 cfm/ft? face openings 
Minimum duct velocity = 3500 fpm 
Ne = 0.25 VPq 


AMERICAN CONFERENCE, 
: OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


METAL SPRAYING 


ie or eo 70] 


10-152 Industrial Ventilation 


10.95 WOODWORKING 


Woodworking equipment generates large amounts of wood 
dust by abrasive or cutting action. It is important to provide 
good ventilation for all equipment as the broad particle size 
distribution of wood dust creates the potential for health and 
housekeeping problems as well as fire hazards. Excessive 
amounts of dust, if allowed to accumulate inside equipment 
and in shop areas, can create fire or explosion hazards. An 
additional consideration should be the toxicity of the wood 
species used. 


In many instances, woodworking equipment, such as saws 
and sanders, generates air flow patterns which make dust 
control difficult. Exhaust hoods should enclose the operation 
as muchas possible. Where the equipment tends to eject wood 
dust (e.g., at sanding belt pulleys) the exhaust hood should be 
placed in the ejection path. 


Enclosures must incorporate cleanout doors to prevent dust 
build-up. Duct velocities should be maintained at a minimum 
of 3500 fpm to prevent settling and subsequent clogging of 
the duct. 


Exhaust flow rates will vary with equipment type and size. 
Design data are provided for a number of operations in 


VS-10-95-1 through 10-95-20 and in Table 10.95.1. Addi- 
tional information for hand-held sanders using Low Volume- 
High Velocity (LV-HV) can be found in sub-section 10.40. 
Where information for a specific operation is not provided, 
data for similar listed operations can be used. 


REFERENCES 


10.95.1 Hamp], V.; Johnston, O.: Control of Wood Dust 
from Horizontal Belt Sanding. American Industrial 
Hygiene Association Journal 46(10)567—-577 
(1985). 


10.95.2 Hampl, V.; Johnston, O.: Control of Wood Dust 
from Disc Sanders. Applied Occupational Hygiene 
6(1 1):938-944 (November 1991). 


10.95.3 Topmiller, J.L.; Watkins, D.A.; Schulman, S.A.; 
Murdock, D.J.: Controlling Wood Dust from Orbital 
Hand Sanders. Applied Occupational Hygiene 
11(9):1131—1138 (September 1996). 


10.95.4 Hampl, V.; Topmiller, J.L.; Watkins, D.S.; Mur- 
dock, D.J.: Control of Wood Dust from Rotational 
Hand-Held Sanders. Applied Occupational Envi- 
ronmental Hygiene 7(4):263—270 (April 1992). 


Specific Operations 10-153 


7 Blade 


re —Hood slotted to 
ag ae enclose blade 


— Hinged door 
for cleanout 


Slotted wood J 
block 


—~—— Top hood 


TOP HOOD DETAIL 


Entire base enclosed 
on all sides 


Blade width, |xhoust flow rate, cfm 
inches | Bottom | Top Total 
Up to 2 350 350 700 
2.to 3 | 350 | 550 900. | 
3 to 4 550 800 1350: . 4} 
4 to 6 550 1100 1650 
6 yo 8 550 1400 

jee, zt i pot A) ane Soc tire, matt a oat 


Minimum duct velocity = 3500 fpm 
he = 1.75 VPy (Point @ in duct riser) 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
f INDUSTRIAL HYGIENISTS 


BAND SAW 


[DATE 72200. | 


FIGURE 


10-154 Industrial Ventilation 


— Saw blade 


2000 fpm 


" taper 


\ 
angle Cleanout door and drawer --—- 


“Saw blade ~ diameter, 7 


inches 


| Exhaust flow rate, 


Up to 16 


16 to 24 _ 


over 24 


Saw with | dado blade 


AMERICAN CONFE 


Minimum duct velocity = 4000 fpm 


h, = 1.78 VP, 


RENCE 


OF GOVERNMENTAL 
INDUSTRIAL HYG ee 


+ 0.25 VP 


d 


FLOOR TABLE SAW 


a - 


—— 1/2" minimum, 


size for 


95—L 


Specific Operations 10-155 


1 3/4 inside diam. 
wow eS oe : 2 flexible hose 


430 cfm 


Blast gate 


Minimum duct velocity = 4000 fpm 
he= 3.5 VPq(point A in duct riser) 


see VS-—80-17 


For booth enclosure, 


AMERICAN CONFERENCE 
/ OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


RADIAL ARM SAW 


MIGURE 


DATE 12 =O) = 


10-156 Industrial Ventilation 


Type hood where table Front view 
is cut through 


\ 
7\ ca 
\ 


le “oN 
{9} 


as 
eas 
a 


Front view 
of hood 


| —/| | 


Type hood where table 
is not cut through 
flow rate 
cfm 


Saw diameier, | Exh. 
inches 


Minimum duct velocity = 4000 fpm 


1.78 VP, + 0.25 VPy 


AMERICAN CONFERENCE peettattass Ae Sia 
| SWING SAW 


| OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


Specific Operations 10-157 


Qg= 100 CFM 


Guard Frame 


Saw Table 


Side View 


AMERICAN CONFERENCE 


OF GOVERNMENTAL 


INDUSTRIAL HYGIENISTS 


End View 


TABLE SAW GUARD EXHAUST 


10-158 Industrial Ventilation 


Drum surface Exhaust flow rate 
square inches cfm 


Up to 200 
(and less than 10° 350 


Minimum duct velocity = 3500 fpm 
Entry loss depends on hood design. 
he= 1.78 VR + 0.25 VRy as illustrated 


AMERICAN CONFERENCE 
: OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


YINGLE DRUM SANDER 


FIGURE 


Specific Operations 10-159 


< Cleanout 


Feed | Note  & 
pias am 

| yi 

- Drum covers necessary. Hinge 
or otherwise provide for 

maintenance. 


Exhaust flow rates 


Total exhaust for 
Drum length machine cfm/drum * 
Up 46 84° 550 
31° to 49” te 790 
49” to 67” 1100 
| over 67) 1400 | 
Brush rolls {350 cfm at brush 


Note: 


Provide one more take off than 
the number of drums. 


Minimum duct velocity = 3500 fpm 
he= 0.25 VFy 


CONFERENCE 
OF GOVERNMENTAL 


AMERICAN 


MULTIPLE DRUM SANDER 


| INDUSTRIAL HYGIENISTS ME ISTO SOS TT 


10-160 Industrial Ventilation 


Disc diameter, Total exhaust flow rate Applies to 
inches cfm duct 


A 


* Two bottom branches. 
** One top and two bottom branches. 


Minimum duct velocity = 3500 fpm 
Entry loss depends on hood design. 
he= 1.0 VPs + 0.25 VFy as illustrated 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


DISC SANDERS 


DATE 1209) FIGURE VS Q5— 12 : 


Specific Operations 10-161 


To—~ Plenum 


Q, (See Note 2) 


Length, L depends 
on size of disk 


a 
eg 


Jet Stripper 
B location, mount 
to work table 


0.625” diam. — 
(see Note 5) 


0.375” diam. 
air inlets 
0.125" diam. to ———— 
pressure gauge 
pressure line optional 


Jet Stripper 
(see note 1) 
NOTES: 
1. Jet stripper may be added as shown. It is used in 

addition to the ventilation specified in VS-95-12. 


2. Qe 18 same as shown in VS—95—12. 

3. Clearance between stripper nozzle and disk is 0.25”. 
4. Nozzle opening diameter is 0.035”. 

5. Total air flow (Q,) to stripper = 1.8 L. (L in inches) 
6. Stripper inlet pressure = 15 psi. 


The Jet Stripper is patented (45,099,616). Use of the devise (not for sale) is 
permissible. To obtain information regarding license for commercial production 


: of the devise contact NIOSH, 1—800—35NIOSH. 


AMERICAN CONFERENCE 
OF GOVERNMENTAL 
f INDUSTRIAL HYGIENISTS 


OPTIONAL JET STRIPPER 
FOR DISK SANDER 


10-162 Industrial Ventilation 


he 
“ \ Tail 


~ Table can raise and lower. 


HORIZONTAL BELT SANDERS 
Belt width, Exhaust flowrate CFM 


inches Head end 
Up to 6 nae 
6 to 9 
9 to 14 | B00 


over 14 —~——S«| «1100 


Minimum duct velocity = 3500 fpm 
he= 0.40 VPy 


AMERICAN CONFERENCE : a aa Ae eae 
OF COVERNMENTAL HORIZONTAL BELT SANDERS 


INDUSTRIAL HYGIENISTS | ST ae 


Specific Operations 10-163 


Main Hood: 
W belt width, in. 
Qe 75 cfm per inch of belt width 


Auxiliary Hood: 
La = auxiliary hood length to extend 3’ — 6° 
into end of sanding crea. 
135 cfm per ft of hood length 


Notes: 
1.Tips of stripper jets must be positioned 0.25” from 
belt and inside hood face. 


2.Keep clearances between hood, belt and workrest 
fo a minimum. 


Auxiliary hood 


4 tubes per inch 


ome) 4 Orifice Orifice 
ue 


WLM” aoe Ww 
L 0.128" x 
0.063" Stripper Jet 


Tube and Orifice Detail 2» Harewidihiy tk 


S 0.5 cfm per tube, 
Reference 10.95.1 inlet pressure 10 — 12 psig 


AMERICAN CONFERENCE HORIZONTAL BELT SANDER 
OF GOVERNMENTAL — | PUSH= POLL. SYS1LEM 


INDUSTRIAL HYGIENISTS fia=——7S—99— FO 


10-164 Industrial Ventilation 


Rear table > 


a Front table 
= 
| 
Velocity at this 


— 2’) minimum clearance 
space, 2000 ae \ 
minimum 


Knife length, Exhaust flowrate 
inches cfm 


_350 


‘over 20 7 


Minimum duct velocity = 4000 fpm 
he= 1.0 VRE + 0.25 VFy 


AMERICAN CONFERENCE 

: / Miter 

OF GOVERNMENTAL JOINTEL: 
ees TRIAL HYGIENISTS 


pave 12-90 DO {riGuRE. VO = OF Or). : 


Specific Operations 10-165 


Z 


lenum exhaust 
copper fitting 1” elbow 
or equivalent. 


Holes to match 
pad mounting holes (Note 4) 


Plenum machined from metal of plastic. 
Plenum can be used with air or electrically powered sander. 
Qe = 50 cfm at approximately 25” wg. 

. Mounting holes located to match sander pad mounting holes. 


The Exhaust Plenum Retrofit has a patent pending. Use of the device (not for 
sale) is permissible. To obtain information regarding the license for commercial 
production of the devise contact NIOSH, 1—800—35NIOSH 


Reference: 10.95.3 


AMERICAN CONFERENCE 
p OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


oh 


EXHAUST PLENUM RETROFIT 
FOR ORBITAL HAND SANDER 


10-166 Industrial Ventilation 


—g§———————-— Sander Air Supply 


——————————— Normal Sander Aspirator 


se 7 4 
Ra 1 : Qe 
Modified Sander Pad 


Note 1 


7 slot depth = 


slot radius = 1/2 d 


1/2 d 


Existing 
Pad Holes 


Modified Sander Pad. 


NOTES: 

1. Additional exhaust plenum to 
fit sander. 

2. Qe= 50 cfm at 25" wg 

3. Qe is in addition to normal sander 
aspirator exhaust. 

4. Additional exhaust may be supplied 

by standard shop vacuum cleaner. 


The Auxiliary Exhaust Modification Configuration is patented (#5,105,858). 
Use of this device (not for sale) is permissible. To obtain information regarding 
license for commercial production of the device contact NIOSH, 1—800—35NIOSH. 


Reference: 10.95.4 B 

CO AUXILIARY EKXHAUST RETROFIT) 

ee a el | FOR AIR POWERED RANDOM | 

: ORBITAL HAND SANDER 
INDUSTRIAL HYGIENISTS 


TABLE 10.95.1. 


Specific Operations 10-167 


Miscellaneous Woodworking Machinery not Given in VS Prints 


The following list of recommended exhaust volumes is for average-sized woodworking machines and is based on many years of experience. It must be noted that 
some modern, high-speed, or extra-large machines will produce such a large volume of waste that greater exhaust volumes must be used. Similarly, some small 
machines of the home workshop or bench type may use less exhaust air than listed. 


Self-feed Table Rip Saw 
Exhaust Flow Rate, cfm 


Saw Diameter, inches Bottom Top 
Up to 16 inclusive 440 350 
Over 16 550 350 
Self-feed, not on table 800 550 


Gang Rip Saws 
Exhaust Flow Rate, cfm 


Saw Diameter, inches Bottom Top 
Up to 24, inclusive 550 350 
Over 24 to 36, incl. 800 440 
Over 36 to 48, incl. 1100 550 
Over 48 1400 550 


Vertical Belt Sanders 
(rear belt and both pulleys enclosed) 
and 
Top Run Horizontal Belt Sanders 


Belt Width, inches Exhaust Flow Rate, cfm 


Up to 6, incl. 440 
Over 6 to 9, incl. 550 
Over 9 to 14, incl. 800 
Over 14 1100 


Swing Arm Sander: 440 cfm 
Single Planers or Surtacers 


Exhaust Flow Rate, cfm 


Up to 20” knives 785 
Over 20” to 26” knives 1100 
Over 26” to 32” knives 1400 
Over 32” to 38” knives 1765 
Over 38” knives 2200 


Double Planers or Surfacers 
Exhaust Flow Rate, cfm 


Bottom Top 
Up to 20” knives 550 785 
Over 20” to 26” knives 785 1100 
Over 26” to 32” knives 1100 1400 
Over 32” to 38” knives 1400 1800 
Over 38” knives 1400 2200 


Molders, Matchers, & Sizers 


Exhaust Flow Rate, cfm 


Total | Size, inches Bottom Top Right Left 
790 | Up to 7, incl. 440 550 350 350 
900 | Over 7 to 12, incl. 550 800 440 440 

1350 | Over 12 to 18, incl. 800 1100 550 550 

Over 18 to 24, incl. 1100 1400 800 800 
Over 24 1400 1770 1100 1100 
Exhaust Flow Rate, cfm 
Total - 
Sash stickers 550 
900 Woodshapers 440 to 1400 

1240 Tenoner Same as moulder 

1650 | automatic lathe 800 to 5000 

2060 | Forming lathe 350 to 1400 

Chain mortise 350 
Dowel machine 350 to 800 
Panel raiser 550 
Dove-tail and lock corner 550 to 800 
Pulley pockets 550 
Pulley stile 550 
Glue jointer 800 
Gainer 350 to 1400 
Router 350 to 800 
Hogs 

Up to 12” wide 1400 

Over 12” wide 3100 
Floorsweep 

6” to 8” diameter 800 to 1400 

Total 

1355 

1885 

2500 

3200 


3600 


10-168 industrial Ventilation 


10.99 MISCELLANEOUS OPERATIONS 


In the previous sections of the chapter, hood ventilation 
sketches were grouped together because they provided venti- 
lation concepts for similar operations, used the same ventila- 
tion approach, or were applicable within the same industry. 
However, not all hood ventilation sketches are so easily 
categorized. 


This section provides a location for those hood ventilation 
sketches that do not fit in other sections. Some have a unique 
application such as VS-99-04 for the Pistol Range. Others 
have such broad application that they could appear in many 
of sections (e.g., the canopy hood in VS-99-03). In other 
cases, this section will be used for new ventilation sketches 
for a particular application or industry. Such sketches will 
reside in this section until other hood ventilation sketches are 
developed and a new section formed. Finally, this section will 
be used for tabular presentation of specific design parameters 
for a variety of operations which could not be adequately 
described in previous sections. 


REFERENCES 


10.99.1 Pennsylvania Department of Labor and Industry: 
Abrasive Wheel Manufacture. Safe Practices Bulle- 
tin No. 13. 

10.99.2 

10.99.3 


Hartzell Propeller Fan Company, Bulletin 1001. 


Goldfield, J.; Brandt, F.E.: Dust Control Techniques 
in the Asbestos Industry. A paper presented at the 
American Industrial Hygiene Conference, Miami 
Beach, FL (May 12-17, 1974). 


Hama, G.M.: Ventilation Control of Dust from Bag- 
ging Operations. Heating and Ventilating (April 
1948). 


Hutcheson, J.R.M.: Environmental Control in the 
Asbestos Industry of Quebec. C 1 MM Bulletin 
64(712):83-89 (August 1971). 

Private Communications. Occupational Health Pro- 
tection Branch, Ontario Ministry of Labour, Ontario, 
Canada (October, 1976). 

Hama, G.; Frederick, W.; Monteith, H.: Air Flow 
Requirements for Underground Parking Garages. 


American Industrial Hygiene Association Journal, 
Vol. 22, No. 6 (December 1961). 


Kane, J.M.: Design of Exhaust Systems. Heating and 
Ventilating 42:68 ( November 1945). 


Oddie, W.M.: Pottery Dusts: Their Collection and 


10.99.4 


10.99.5 


10.99.6 


10.99.7 


10.99.8 


10.99.9 


10.99.10 


10.99.1] 


10.99.12 


10.99.13 


10.99.14 


10.99.15 


10.99.16 


10.99.17 


10.99.18 


10.99.19 


10.99.20 


10.99.21 


10.99.22 


10.99.23 


Removal. Pottery Gazette 53:1280 (1928). 


B.F. Sturtevant Company: What We Make. Catalog 
No. 500. 


Brandt, A.D.: Industrial Health Engineering, John 
Wiley and Sons, New York (1947). 


Kane, J.M.: Foundry Ventilation. The Foundry (Feb- 
ruary and March 1946). 


Kane, J.M. Foundry Ventilation. University of 
Michigan In-Service Training Course (October 
1945). 


Fen, O.E.: The Collection and Control of Dust and 
Fumes from Magnesium Alloy Processing. Peters- 
Dalton, Inc. January 1945). 


Postman, B.F.: Practical Application of Industrial 
Exhaust Ventilation for the Control of Occupational 
Exposures. American Journal of Public Health, 
30:149 (1940). 


New York Department of Labor: Rules Relating to 
the Control of Silica Dust in Stone Crushing Opera- 
tions. Industrial Code Rule No. 34 July 1942). 


DallaValle, J.M.: Exhaust Hoods. Industrial Press, 
New York (1946). 


Hatch, T.; et al: Control! of the Silicosis Hazard in 
Hard Rock Industries. I]. An Investigation of the 
Kelley Dust Trap for Use with Pneumatic Rock 
Drills of the Jackhammer Type. Journal of Industrial 
Hygiene 14:69 (February 1932). 


Hay, Capt., P.S.: Modified Design of Hay Dust 
Trap. Journal of Industrial Hygiene 12:18 (January 
1930). 


Riley, E.C.; et al: How to Design Exhaust Hoods for 
Quartz-Fusing Operations. Heating and Ventilating, 
37:23 (April 1940). 


Riley, E.C.; DallaValle, J.M.: A Study of Quartz- 
Fusing Operations with Reference to Measurement 
and Control of Silica Fumes. Public Health Reports 
54:532 (1939). 


Yaglou, C.P.: Ventilation of Wire Impregnating 
Tanks Using Chlorinated Hydrocarbons. Journal of 
Industrial Hygiene and Toxicology 20:401 (June 
1938). 


American Air Filter Co., Inc.: Usual Exhaust Re- 
quirements (for) Grain Elevators, Feed and Flour 
Mills. Louisville, KY (April 1956). 


Specific Operations 10-169 


~ Flexible connection if desired 


SS _--» Top take-off preferred 


-Complete enclosure 


FLAT DECK SCREEN 


Q =200 cfm/ ff through hood openings, but not fess thon 
50 cfm/fi~ screen area. No increase for multiple decks 
Minimum duct velocity = 3500 fom 


hg = 0.90 Vey 


Complete 
enclosure 


Screen ~ / CYLINDRICAL SCRIEION 


= 100 cfm/ft* circular cross Sgction of 
screen; at least 400 cfm/ft* of 
enclosure opening 
Minimum duct velocity = 3500 fpm 


he = 0.50 VP, 


Oversize 


AMERICAN CONFERENCE og 


} OF GOVERNMENTAL een 
J INDUSTRIAL HYGIENISTS | FSO Yan Von T 


10-170 Industrial Ventilation 


2 
Q = 50-100 cfm/ft of table top. 
Minimum duct velocity = 2000 fom 
he = 1.78 VPs + 0.25 VRy 


Note: See “Open Surface Tanks’, VS-70-01 and VS—70-02 
for other suitable slot types. Air flow rate may 
be calculated on dilution basis if data is available. 
Maximum plenum velocity = 1/2 slot velocity. 
Large plenum essential for good distribution. 


AMERICAN CONFERENCE | Agee estate 
OF GOVERNMENTAL — | pee 
INDUSTRIAL HYGIENISTS 


Specific Operations 


45 Minimum 


PROCESS 


Not to be used where material is toxic and worker must bend 
over tank or process. 


Side curtains are necessary when cross~drafts are present. 


for open type canopy. 
P = perimeter of tank, feet. 
V = 50-500 fpm. See 


for two sides adjacent enclosed. 
W & L are open sides of hood. 
V = 50-500 fpm. See Chapter 3 


for three sides enclosed. (booth) 
V = 530-500 fpm. See Chapter 3 


he= 0.25 VPy 
Duct velocity = 1000-3000 fpm 


AMERICAN CONFERENCE 
| OF GOVERNMENTAL 
| INDUSTRIAL HYGIENISTS 


CANOPY HOOD 


10-171 


10-172 Industrial Ventilation 


ce If baffle is used, it must be located at 
the extreme down range end of the booth. 


a ~ lf baffles are present use a 
~ 8-10? ceiling height bar positioned to provide a 
6 inch space between shooter 
and bench. 


a “ be avoided. 


iz Double 0.25° pegboard wall will provide better distribution 
than single. Wall must be constructed with spacers to 
assure free air circulation within wall. 


Q(minimum) = 50 HW, but not less than 20 cfm/fft 
of room cross-sectional area 


Notes: 
Replacement air distribution: 
Uniform air distribution necessary. 
Perforated rear wall or ceiling plenum system 
preferred. Grilles and diffusers are not 
recommended. 


NIOSH certified dust respirator for lead is necessary 
during clean-up and lead removal from bullet trap. 


Acoustical material on walls, ceiling and thick fabric 
on bench top are recommended 


INDOOR PISTOL AND~ 
SMALL BORE RIFLE 
RANGE VENTILATION 


AMERICAN CONFERENCE 
: OF GOVERNMENTAL 
INDUSTRIAL HYGIENISTS 


[pate 7>— 90 ~*Y[ricuRE VS “99-07 I 


Specific Operations 10-173 


- Slots 


Side baffles desirable 


- 6 freeboard in tank 


E 45° minimum. slope- 


AS 


Use at least two slots. One at 


botiom of hood. 


Q = 150 re bed (150LW) 
Slot velocity = 2000 fpm 

he= 1.78 VP. + 0.25 VPy 
Minimum duct velocity = 3500 fpm 
W not to exceed 36° 


Free board must be maintained to prevent material carryout. 


| AMERICAN CONFERENCE | 
| OF GOVERNMENTAL 
| INDUSTRIAL HYG 


PLUIDIZED BEDS 


FIGURE 


10-174 Industrial Ventilation 


Q = 200 L x W 

he = 1.78 slot VPs + 0.25 duct VPq 
Duct velocity = 2500 — 3000 fpm 
Slot size for 1500 — 2000 fpm 


AMERICAN CONFERENCE 


OF GOVERNMENTAL OUTBOARD MOTOR TES 


INDUSTRIAL HYGIENISTS fie 


Specific Operations 10-175 


pore Table with hood 


in place 


A. To Exhaust 


Hood 
Table 


Cross—section of 
table and hood 


Qrom = Q;+ Q2,= 725 cfm 
Slot Velocity = 2000 fpm 
Duct Velocity = 2000 fpm 
he = 1.78 VPS + 0.5 VPy 


Note 1: Position of exhaust take-offs may vary due to table configuration 
place to assure same airflow distribution into slot. 


_ Reference 10.99.24 

| AMERICAN CONFERENCE 
| OF GOVERNMENTAL 

| INDUSTRIAL HYGIENISTS 


MORTUARY TABLE 


10-176 Industrial Ventilation 


Slot hood ~~ 
ae pe Baf‘le — 


Note 1 


—-- Slot hood 


Perforated plate -~ ) 
™. is oye O° 

26) 1° 2 

8 aa ° OS. a ee Note | 

20% 0080 | Perforated plate [ 
10% open area ff 


1/2’ hole 


Tank drain 


45 cfm/ft? of tank area 


= 65 cim/tt? of tank grea 


Plenum velocity < 1/2 slot velocity 
Slot velocity = 2000 fom 


Slide damper 7 _ 
(full width of hood) he= 1.78 Vy +.25 VPs 


: Build perforated plate in 
sections for easy cleaning 
Allow freeboard height al least 9° 


: Boffle to extend to ends of tank 
6° (minimum) above top of workpiece 


: Adjust slot to achiev